JP2004257991A - Molecular calculation apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for utilizing the parallelism of molecular operation and calculating faster than the conventional electronic computer. <P>SOLUTION: A calculation apparatus for implementing an information processing method using a nucleic acid is provided. The calculation apparatus is a molecular calculation apparatus provided with an electronic calculation part and a molecular calculation part. The electronic calculation part substantially controls the function of the molecular calculation part, and a molecular calculation apparatus implements the operation by the molecule under the control of the electronic calculation part. The molecular calculation part substantially implements the calculation by implementing a chemical reaction of the molecule. The calculation apparatus for efficiently implementing data parallel processing and gene analysis by implementing the calculation using the nucleic acid molecule is disclosed. An incredibly large memory capacity and an incredibly high parallel processing capacity are achieved as compared with the conventional computer by expressing data and a program by the nucleic acid molecule, substituting the molecular reaction for the operation defined by the program and implementing the reaction. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【０１２８】
ここで、記号の「∧」は論理積、「∨」は論理和、「¬」は否定を表す。
【０１２９】
【数２】

【０１３０】
本装置を用いて４変数１０節の３ＳＡＴ問題を解いた。問題を数式１に、問題の解を数式２に示す。数式１の問題の論理式はｘ_１、ｘ_２、ｘ_３、ｘ_４の４つの変数からなり、３ラテラルの節が１０個、論理積で結合されている。この式を満たす変数の値の組、すなわち解が有るかどうか、有るならばその変数の値の組を求めるのが目的である。
【０１３１】
【数３】

【０１３２】
この問題を解くために数式３に示すプログラムを本装置で実行する。記述法はパスカルに準じている。具体的には、数式３に示すプログラム（ｄｎａ３ｓａｔ）を原始プログラムとして図１の入力部１１で入力する。入力された原始プログラムは、翻訳などを経て制御命令手順テーブルが生成され、分子計算部２２で分子計算が実行される。そして、（ｓ２）〜（ｓ８）のステップを経て数式３に示す計算結果が得られる。
【０１３３】
関数ｄｎａ３ｓａｔが、問題を与えたときに解が存在した場合に解を出力するメイン関数である。この関数ｄｎａ３ｓａｔ中には関数ｇｅｔｕｖｓａｔが組み込まれている。これら関数ｄｎａ３ｓａｔと関数ｇｅｔｕｖｓａｔはｇｅｔ、ａｍｐｌｉｆｙ、ａｐｐｅｎｄ、ｍｅｒｇｅ、ｄｅｔｅｃｔの５つの基本関数で構成されている。基本関数のうちｇｅｔ、ａｍｐｌｉｆｙ、ａｐｐｅｎｄは、原始プログラムの手続または関数として図１５から図１８に示す化学反応により実行される。
【０１３４】
次に各基本関数に対応付けられた化学反応、化学反応に付随する操作を説明する。最初に、図１５に従ってｇｅｔ（Ｔ，＋ｓ）、ｇｅｔ（Ｔ，−ｓ）関数について述べる。
【０１３５】
関数ｇｅｔは、オリゴヌクレオチドの混合溶液Ｔ（ｔｕｂｅ）から、ｓなる配列を含む１本鎖のオリゴヌクレオチドまたは該配列を含まないオリゴヌクレオチドを取得する反応操作を符号分子の演算反応として定義した関数である。ここで、ｓとは数塩基または数十塩基の特定の配列を示している。ｇｅｔ（Ｔ，＋ｓ）はｓ配列を含むオリゴヌクレオチドを、ｇｅｔ（Ｔ，−ｓ）はｓを含まないものを取得する。Ｔ，＋ｓ，−ｓは、分子を核酸配列に基づき符号化した符号分子に対応付けて定義される変数である。
【０１３６】
まず、ｓに相補的な配列を持ち５’端ビオチンを標識したオリゴヌクレオチドをＴに入れ、ｓ配列を備えたオリゴヌクレオチドとアニーリングさせる。
【０１３７】
このアニーリングでできたハイブリッドを、例えば、ストレプトアビジンを表面に結合した磁気ビーズで捕獲する。このあとｓ配列に相補的なオリゴヌクレオチドのなすハイブリッドが解離しない温度で磁気ビーズの緩衝溶液で洗浄する（即ち、コールドウォッシュする）。これにより、ｓ配列を持たないオリゴヌクレオチドを緩衝液から取得でき、ｇｅｔ（Ｔ，−ｓ）が実行できる。
【０１３８】
また、関数ｇｅｔ（Ｔ，＋ｓ）の実行は、コールドウォッシュが終わってから、ハイブリッドが解離するような比較的高温の緩衝溶液で磁気ビーズを洗えばよい。これにより、その緩衝溶液からｓ配列を持つオリゴヌクレオチドを取得できる。即ち、１つの操作で２つの関数が実行できる。
【０１３９】
次に、図１６に従って関数ａｐｐｅｎｄ（Ｔ，ｓ，ｅ）の反応について述べる。ｅは、分子を核酸配列に基づき符号化した符号化分子に対応付けて定義される変数である。
【０１４０】
この関数は、オリゴヌクレオチドの混合液Ｔの中でｅ配列をその３’端に備えた１本鎖ＤＮＡの３’端に、ｓなる配列を備えたオリゴヌクレオチドをライゲーションにより連結し、そのｓ配列が連結された１本鎖のオリゴヌクレオチドを取得する反応操作を符号分子の演算反応として定義した関数である。ここで、ｓ、ｅとは先に述べたのと同様に数塩基若しくは数十塩基の特定の配列を指す。
【０１４１】
この関数ａｐｐｅｎｄ（Ｔ，ｓ，ｅ）の反応では、次のオリゴヌクレオチドを用いる。
【０１４２】
ｓ配列のオリゴヌクレオチドは、この５’端がリン酸化されている。また、連結オリゴヌクレオチドは、５’端にビオチン標識されており、５’端側のｓに相補的な配列と３’端側のｅに相補的な配列とを隣接して含有する。これらをチューブＴに入れて反応を始めると連結オリゴヌクレオチドは、Ｔ中の標的オリゴヌクレオチド配列ｅとｓ配列オリゴヌクレオチドとでハイブリッドを形成する。このハイブリダイゼーション反応は、ミスマッチのハイブリッドが形成されないように比較的高温で行い、Ｔａｑライゲースなど高温で活性の高い酵素を用いて連結される。このあとストレプトアビジン磁気ビーズなどによりハイブリッドを捕獲し、ハイブリッドが解離するような高温で洗えば（即ち、ホットウォッシュ）、３’末端にｓ配列が連結されたオリゴヌクレオチドを回収することができる。
【０１４３】
次に図１７に従って、ａｍｐｌｉｆｙ（Ｔ、Ｔ_１、．．．．Ｔ_ｎ）関数の反応について述べる。このａｍｐｌｉｆｙ（Ｔ、Ｔ_１、．．．．Ｔ_ｎ）関数の反応は、反応溶液Ｔに含まれるオリゴヌクレオチドをＰＣＲ反応により増幅し、増幅され二本鎖になったオリゴヌクレオチドを１本鎖に変えてから、Ｔ_１、．．．Ｔ_ｎなるｎ個の反応溶液に分割する反応である。本実施例ではＴに含まれるオリゴヌクレオチドの両端には共通な増幅用の配列が具備され、１組のプライマーにより全てのＴ中のオリゴヌクレオチドを増幅できるようにしている。プライマーのうち、増幅対象のオリゴヌクレオチドの３’端に相補的な配列をもつものの５’端にはビオチン標識が施されている。Ｔのオリゴヌクレオチドをこの共通プライマーにより増幅し、ストレプトアビジン磁気ビーズにより捕獲する。これを９４℃などの高温で全ての２本鎖が解離するように洗浄すれば、もともとＴに含まれていた側の鎖が緩衝溶液中に抽出できる。この抽出溶液を均等にＴ_１、．．．Ｔ_ｎのｎ個の反応溶液に分割すればよい。
【０１４４】
関数ｍｅｒｇｅは、引数の複数の溶液を１つの溶液に纏める反応操作を符号分子の演算反応として定義した関数である。
【０１４５】
最後に関数ｄｅｔｅｃｔを図１８に従って説明する。検出は、手動による反応で実施することも可能である。関数ｄｅｔｅｃｔを本装置により行う方法の例を以下に示す。ここではｇｒａｄｕａｔｅｄ ＰＣＲなる、エーデルマンの論文（Ｓｃｉｅｎｃｅ， ２６６， １０２１−４）において開示された検出手段を用いて行う例を示す。
【０１４６】
図１８において、本装置で生成される解を示す核酸分子は、５’端から順に４つの変数が真（即ち、Ｔ、ｔｒｕｅ）または偽（即ち、Ｆ、ｆａｌｓｅ）を表す配列が連結した配列を有す。この解の配列は、同じく図１８に示す全ての可能な解を検出できるようなプライマーにより個別にＰＣＲ増幅する。図中、配列名の上に引いた横線は、相補配列を示す。仮に、図１８の最初に示すような配列の解が得られていると、図１８の最後に示すプライマーの組で、且つ図示した長さのＰＣＲ産物が得られる。これら産物の存在と長さをゲル電気泳動にて確認すれば、ＰＣＲ産物の得られたプライマーから解の配列が明らかになる。なお、関数ｄｅｔｅｃｔは、ＤＮＡチップによる出力も含まれる。
【０１４７】
以上の関数を表１に纏めた。
【０１４８】
【表１】

【０１４９】
上述の数式３に示すプログラムの主たる関数はｄｎａ３ｓａｔである。その中に先に述べた基本関数と、基本関数からなる関数ｇｅｔｕｖｓａｔが組み込まれている。このように、基本関数を組み合わせて新たに手続または関数を定義し、この手続または関数を用いたプログラムを作成することが可能である。最初にプログラム上の変数名の説明を行う。
【０１５０】
Ｔとはチューブ（ｔｕｂｅ）の略であり、Ｔ_２とはｘ_１、ｘ_２の２変数の取りうる全ての真偽の組４つを表すオリゴヌクレオチドを含む溶液である。Ｘ_１ ^Ｔとは、変数ｘ_１の値が真であることを示すオリゴヌクレオチドで長さは２２塩基である。Ｘ_２ ^Ｆとは、変数ｘ_２の値が偽であることを示すオリゴヌクレオチドであり、長さは同じく２２塩基である。よってＴ_２に含まれるオリゴヌクレオチドの１つ、Ｘ_１ ^ＴＸ_２ ^Ｆは４４塩基の長さの、ｘ_１＝１、ｘ_２＝０なる割り当てを表す１本鎖のオリゴヌクレオチドである。ｊ、ｋはループの進行状態を表す整数で、ｊは問題の論理式の何節目を計算しているかを示し、ｋは何番目の変数を計算しているかを表す。ｕ、ｖ、ｗは問題の論理式の各節内のひとつひとつのリテラルを示す。各リテラルは、数式１の問題であれば、ｕ_１＝ｘ_１、ｖ_２＝ｘ_２、ｗ_１＝¬ｘ_３となる。配列名Ｘの上にあるバーは相補配列を意味する。また、Ｘの右肩のＴ／ＦはＴおよびＦを指し、Ｘ^ＴとＸ^Ｆを意味する。
【０１５１】
以下、数式３のプログラムを図２０のフローチャートに沿って順を追って説明する。実際にはプログラム解釈部がこのプログラムの関数および条件分岐、ｆｏｒループを一連の実験操作に展開して実験計画を立てる。最初に問題の理論式はプログラムでの処理を容易にするために、各節で変数の添え字の昇順にリテラルを並べ替えておく。また、２変数ｘ_１、ｘ_２に可能な割り当てを全て行った溶液Ｔ_２を準備しておく。
【０１５２】
関数ｄｎａ３ｓａｔにおいて、最初のｆｏｒループのｋ＝３、即ち、ｘ_３に割り当てる値を決定するループについて説明する。最初にａｍｐｌｉｆｙ関数でＴ_２を増幅する。この場合、増幅後、Ｔ_２と同じオリゴヌクレオチドを持つＴ_ｗ ^Ｔ、Ｔ_ｗ ^Ｆに分注し、計３本の溶液チューブを得る。次にｋのループ内のｊのｆｏｒループに入る。ここでは、論理式の第ｊ番目の節の充足性を評価する。第１の条件分岐で、ｊ＝１では第１節の第３リテラルｗ_１がｘ_３と等しいかを調べる。これは、電子計算部２１で予め論理式を調べ、ｔｈｅｎ以降を実行するかどうかを判断し、実験計画に入れるかを決定しておく。ループ内第２の条件分岐で第１節の第３リテラルｗ_１が¬ｘ_３であるのでｔｈｅｎ以降を実行する。
【０１５３】
関数ｇｅｔｕｖｓａｔは下段に別に定義されており、それぞれの引数、Ｔに対しては最初にａｍｐｌｉｆｙで生成したＴ_ｗ ^Ｔチューブ（内容および濃度はＴ_２と同じ）、ｕに対してはｕ_１（＝ｘ_１）、ｖに対してはｖ_１（＝ｘ_２）を入力する。最初のＴ_ｕ ^ＴはＴ_ｗ ^Ｔの中からＸ_１ ^Ｔを含むオリゴヌクレオチドを抽出して溶液を生成する。これにより、Ｔ_ｕ ^Ｔは｛Ｘ_１ ^ＴＸ_２ ^Ｔ、Ｘ_１ ^ＴＸ_２ ^Ｆ｝という内容になる。次にＴ’_ｕ ^ＦはＴ_ｗ ^Ｔの中からＸ_１ ^Ｔを含まないオリゴヌクレオチドを抽出して溶液を生成する。これによりＴ’_ｕ ^Ｆは｛Ｘ_１ ^ＦＸ_２ ^Ｔ、Ｘ_１ ^ＦＸ_２ ^Ｆ｝という内容になる。３つ目にＴ_ｕ ^Ｆは、Ｔ’_ｕ ^Ｆの中からＸ_１ ^Ｆを含むオリゴヌクレオチドを抽出して溶液を生成する。これによりＴ_ｕ ^Ｆ＝｛Ｘ_１ ^ＦＸ_２ ^Ｔ、Ｘ_１ ^ＦＸ_２ ^Ｆ｝という内容になる。ここでのこの行は無視可能というコメントがあるが、生成される溶液の内容を見れば明白である。実験上エラーが生じることのないように加えた行であるので、念のための実行する。無視した場合は、前行のＴ’_ｕ ^ＦをＴ_ｕ ^Ｆと読み替える。４番目のＴ_ｖ ^ＴはＴ_ｕ ^Ｆの中からＸ_２ ^Ｔを含むオリゴヌクレオチドを抽出して溶液を生成する。これにより、Ｔ_ｖ ^Ｔ＝｛Ｘ_１ ^ＦＸ_２ ^Ｔ｝という内容になる。最後に関数ｍｅｒｇｅにより、Ｔ_ｕ ^ＴとＴ_ｖ ^Ｔが混合されてＴ^Ｔが生成され、ｒｅｔｕｒｎにより出力される。Ｔ^Ｔは｛Ｘ_１ ^ＴＸ_２ ^Ｔ、Ｘ_１ ^ＴＸ_２ ^Ｆ、Ｘ_１ ^ＦＸ_２ ^Ｔ｝となりもとのｄｎａ３ｓａｔ関数では溶液Ｔ_ｗ ^Ｔと名を代えて扱うことになる。Ｔ^Ｔの内容は第１節の第１、第２のリテラルがｘ_１∨ｘ_２なる形であるので、第２リテラル¬ｘ_３がいかなる値であろうとも第１節の第１節が真であるｘ_１、ｘ_２への値の割り当ての組を示している。以上により、関数ｇｅｔｕｖｓａｔは既に２つのリテラルに値が割り当てられている節について、３つ目のリテラルの変数に何を割り当ててもその節が真であるような割り当てを表すオリゴヌクレオチドを選別する関数である。
【０１５４】
次に、再びｄｎａ３ｓａｔ内のｋ＝３ループでｊを１つ増加させ、ｊ＝２で、第２節の計算に移る。この節では第３リテラルは第１節と同様に¬ｘ_３なので条件分岐に効いてくるのは下段のｉｆ文である。第１節と同様にｇｅｔｕｖｓａｔを実行する。注意すべきは先に行ったｊ＝１でＴ_ｗ ^Ｔを生成していることである。また、ｖ_２はｖ_２＝¬ｘ_２なる否定が入ったリテラルなので、ｇｅｔｕｖｓａｔ関数のＴ_ｖ ^Ｔを求める式においてＸ_ｖ ^Ｔは、Ｘ_２ ^Ｆと読み替えて関数ｇｅｔを実行する。こうすれば、ｊ＝２で新たに得られるＴ_ｗ ^Ｔは、｛Ｘ_１ ^ＴＸ_２ ^Ｔ、Ｘ_１ ^ＴＸ_２ ^Ｆ｝となる。
【０１５５】
以降、順にｊ＝１０まで実行する。途中、例えば、５節目以降のように３番目のリテラルがｋ＝４である場合は、ｊの値をインクリメントし、ループの次の処理を行う。ｋ＝３で１０節分のループが終了すれば、Ｔ_ｗ ^Ｔの中の３’末端がＸ_２のオリゴヌクレオチドに真を表すＸ_３ ^Ｔをａｐｐｅｎｄする。また、Ｔ_ｗ ^Ｆ中の３’末端がＸ_２のオリゴヌクレオチドに偽を表すＸ_３ ^Ｆをａｐｐｅｎｄする。このあとａｐｐｅｎｄ後の溶液をｍｅｒｇｅし、ｋ＝４のループに移る。ｋ＝４のループが終了すれば、最終的にループを脱し、Ｔ_４チューブを関数ｄｅｔｅｃｔで処理する。
【０１５６】
以上のように計算を実行するので、変数の数をｎ、節の数をｍとすると各基本コマンドの実行回数は、
（ｎ−２）×（ａｍｐｌｉｆｙ＋２×ａｐｐｅｎｄ＋ｍｅｒｇｅ）＋ｍ×（３×ｇｅｔ＋ｍｅｒｇｅ）
である。即ち、変数の数と節の数にほぼ比例した時間で３ＳＡＴ問題を解くことができる。
【０１５７】
また、本発明において、ｄｅｔｅｃｔ部分は、必ずしも反応結果を測定することを意味せず、反応結果が電子計算部へ利用可能に情報伝達されるか、或いは利用者が利用できる形態に変化または抽出された状態に特徴化されていればよい。また、ｄｅｔｅｃｔ部分は測定可能な状態に特徴化されていればよく、測定を人が行ってもよい。従って、本発明において、分子計算部は、反応が終了すると共に利用者が利用できるような生成物または成果物を提供するところまでを実施すればよい。利用者は、本発明の装置により提供された生成物または成果物を種々の目的、例えば、診断、治療、創薬、学術的研究、生物学的データベースの構築、生物学的情報の解読等へと最も効率よく利用することができる。
【０１５８】
【実施例】
分子計算
上述で説明した数４のプログラムを本発明の分子計算装置を用いて実行した。装置の構成を以下に記す。該プログラムを実行した装置の構成を示す。本装置はプレシジョン・システム・サイエンス（ＰＳＳ）社の核酸自動抽出装置ＳＸ８Ｇを改造して製作した。本装置は、制御のためのインテル社製ＰｅｎｔｉｕｍＩＩＩＣＰＵ（登録商標）を搭載した、ＯＳがＷｉｎｄｏｗｓ９８（登録商標）のコンピュータ（電子計算部２１に対応）と、演算反応を実際に行うための試薬槽と反応槽、ＸＹＺの位置制御のできるピペッタと、予備のピペッタ用チップ、温度をコンピュータで制御できるサーマルサイクラ（ＭＪ Ｒｅｓｅａｒｃｈ社製のＰＴＣ−２００）からなる反応ロボット（分子計算部２２に対応）からなる。分子計算部２２の上から見た反応用部材の配置を図２に示す。
【０１５９】
これらの反応容器の間で溶液の受け渡しが行われる。ストレプトアビジン磁気ビーズによるオリゴヌクレオチドの抽出は、ピペッタの特殊チップとチップに近づけたり離したりできる永久磁石により行う。ただし、以上のシステムでは実行できなかった２つの操作、即ち１μＬ程度の微量の酵素分注は人力で行い、関数ｄｅｔｅｃｔでのキャピラリゲル電気泳動は、ベックマン・コールター社製Ｐ／ＡＣＥ−５５１０キャピラリ電気泳動システムにより実行した。
【０１６０】
試薬類とその初期配置を述べる。リガーゼ酵素は先に述べたようにマニュアルで分注するので装置外で保持した。分注はギルソン社製のピペットマン２μＬのピペッタを用いた。
【０１６１】
Ｘで表される各変数の値を示すオリゴヌクレオチドの長さは２２塩基である。また、ここでＸ_１ ^ＴＸ_２ ^Ｆと記した場合５’端側から順にＸ_１が真であることを表すオリゴヌクレオチド、Ｘ_２が偽であることを示すオリゴヌクレオチドの配列をもつ１本鎖オリゴヌクレオチドであることを示す。また、ここで、配列名が「［］」に挟まれている場合は元の配列に相補的な配列を意味する。このため、例えば［Ｘ_１ ^ＴＸ_２ ^Ｆ］ならばＸ_１ ^ＴＸ_２ ^Ｆに相補的なので、実際の配列は５端側からＸ_２ ^Ｆに相補的な配列、Ｘ_１ ^Ｔに相補的な配列の順になっている。
【０１６２】
Ｔ_２溶液（Ｘ_１ ^ＴＸ_２ ^Ｔ、Ｘ_１ ^ＦＸ_２ ^Ｔ、Ｘ_１ ^ＴＸ_２ ^Ｆ、Ｘ_１ ^ＦＸ_２ ^Ｆそれぞれ５ｐｍｏｌをライゲーション用緩衝溶液２０μＬに溶解した）をＭＴＰ３５（即ち、マルチタイタープレート３５）に保持した。ライゲーション反応用緩衝溶液、ストレプトアビジン磁気ビーズ、Ｂ＆Ｗ溶液（ＴＥ溶液にＮａＣｌを加えて、ＮａＣＩが１Ｍなる濃度にした溶液を以降このように呼ぶ。この液はストレプトアビジン磁気ビーズにビオチン化オリゴヌクレオチドを捕獲したり、ハイブリダイゼーション反応を行うときに用いる）は図２のＭＴＰ３７に保持した。
【０１６３】
また、５’端がビオチン標識されているビオチン化オリゴヌクレオチド［ｂＸ_１ ^Ｔ］、［ｂＸ_２ ^Ｔ］、［ｂＸ_３ ^Ｔ］、［ｂＸ_４ ^Ｔ］、［ｂＸ_１ ^Ｆ］、［ｂＸ_２ ^Ｆ］、［ｂＸ_３ ^Ｆ］、［ｂＸ_４ ^Ｆ］をぞれぞれ１０ｐｍｏＬずつＢ＆Ｗ溶液２０μＬに溶解したもの、５’端がリン酸化されているａｐｐｅｎｄオリゴヌクレオチドｐＸ_３、ｐＸ_３ ^Ｔ、ｐＸ_３ ^Ｆ、ｐＸ_４ ^Ｆ、ｐＸ_４ ^Ｔ、それぞれ１０ｐｍｏＬをライゲーション緩衝溶液２０μＬに溶解したもの、５’端にビオチン標識した連結オリゴヌクレオチド、［ｂＸ_２ ^ＴＸ_３ ^Ｔ］、［ｂＸ_２ ^ＦＸ_３ ^Ｔ］、［ｂＸ_２ ^ＴＸ_３ ^Ｆ］、［ｂＸ_２ ^ＦＸ_３ ^Ｆ］、［ｂＸ_３ ^ＴＸ_４ ^Ｔ］、［ｂＸ_３ ^ＦＸ_４ ^Ｔ］、［ｂＸ_３ ^ＴＸ_４ ^Ｆ］、［ｂＸ_３ ^ＦＸ_４ ^Ｆ］それぞれ１０ｐｍｏｌをライゲーション緩衝溶液２０μＬに溶解したものをＭＴＰ３５に配置する。これらのオリゴヌクレオチド溶液および緩衝溶液などは分子計算機動作中は室温で保持した。リガーゼ、関数ｄｅｔｅｃｔで用いるＰＣＲ用ポリメラーゼは装置の外で氷中で、ＰＣＲ反応用緩衝溶液は室温で維持した。
【０１６４】
それぞれの関数に対応する反応での装置の動作を順を追って説明する。
【０１６５】
（ａ）関数ａｍｐｌｉｆｙ
関数ａｍｐｌｉｆｙは、左端の引数の溶液をテンプレートとし、ＰＣＲ増幅して元の溶液の濃度に維持して複数の溶液に分割する反応操作を符号分子の演算反応として定義した関数である。厳密にはもとの溶液のオリゴヌクレオチド濃度を測定してからでなければ増幅できないが、実際には最初のＴ_２で各オリゴヌクレオチドが５ｐｍｏｌで２０μＬの溶液に溶解しているので、２０μＬの溶液に各オリゴヌクレオチドが数ｐｍｏ１溶解しているようにする。即ち、１溶液１オリゴヌクレオチドあたり２ｐｍｏｌとするならば、ＰＣＲ反応液の組成は次の通りである。
【０１６６】
ポリメレース酵素（宝酒造）０．５μＬ（２．５Ｕ）
増幅ＤＮＡの溶液 １ｆｍｏｌ程度の各々のオリゴヌクレオチドを含む
ｄＮＴＰ混合溶液 ８μＬ（２．５ｍＭ、付属品）
反応バッファ １０μＬ（１０倍希釈で使用、付属品）
プライマー 各オリゴヌクレオチドにｆｏｒｗａｒｄ、ｒｅｖｅｒｓｅ側とも５ｐｍｏｌずつ準備
滅菌蒸留水 全液量が１００μＬになるように加えた
増幅には宝酒造のＰｙｒｏｂｅｓｔ^ＴＭ ＤＮＡ ＰｏｌｙｍｅｒａｓｅのＰＣＲ増幅キットを用いた。反応温度条件は、以下の通りである。
【０１６７】
即ち、
１． ９５℃ ３０秒
２． ５０℃ ３０秒
３． ７２℃ ６０秒 １〜３を３０サイクル
である。
【０１６８】
ＰＣＲ反応液の量は分割する溶液の数に応じて変えた。ＰＣＲの後、反応液からＰＣＲ産物をストレプトアビジン磁気ビーズ（ロシュダイアグノスティクス）で捕獲する。磁気ビーズ原液５０μＬ（０．５ｍｇ）をとり、ここから磁石により磁気ビーズのみを抽出して液をＢ＆Ｗ溶液５０μＬに置換する。この磁気ビーズ液とＰＣＲ反応液５０μＬとを混ぜてＰＣＲ産物を捕獲する。捕獲した後、溶液をＢ＆Ｗ溶液５０μＬに置換し、８８℃まで昇温して１本鎖オリゴヌクレオチドを解離させて抽出した。
【０１６９】
（ｂ）関数ｇｅｔ
ｇｅｔ（Ｔ，＋Ｓ）とｇｅｔ（Ｔ，−Ｓ）は一連の操作で同時に行う。
【０１７０】
（１）抽出する溶液Ｔを５０μＬ準備しサーマルサイクラ３８に吐出した。
【０１７１】
（２）Ｔに捕獲する配列の相補鎖をビオチン化したオリゴヌクレオチドを２０ｐｍｏｌを含むＢ＆Ｗ溶液５０μＬをさらに加えて最初のハイブリダイゼーション反応を行った（下記反応温度条件の１，２）。
【０１７２】
ピペッタは各ＭＴＰから液をとり反応を進める。反応温度条件は以下の通りである。即ち、
１． ９５℃ １分
２． ２５℃ １０分（１から２までは１０℃／分で温度を下げる）
３． ５６℃ ３分（２から３までは１０℃／分で温度を上げる）
４． ７５℃ ３分（３から４までは１０℃／分で温度を上げる）
であり、ピペッティングの作業時間を確保しながらサーマルサイクラを制御した。
【０１７３】
（３）２の温度の途中で磁気ビーズ原液５０μＬ分を分散したＢ＆Ｗ溶液５０μＬを混合し、磁気ビーズにビオチン化オリゴヌクレオチドのハイブリッドを捕獲した。磁気ビーズは再びサーマルサイクラー９６穴ＭＴＰ３８に保持した。
【０１７４】
磁気ビーズ液は装置のピペッタに一度吸い込まれた後、ピペットのチップの中の空洞に保持される。このとき液を保持したチップにピペッタに取り付けた移動可能な永久磁石を接近させてビーズを集める。この間にピペットから排液して新たな溶液を吸引すれば溶液の置換や、２本鎖核酸の相補鎖の分離を行うことができる。磁気ビーズを溶液に分散するためには永久磁石を離して数回ピペッティングを行えば十分に撹拌されて磁気ビーズは溶液に分散する。
【０１７５】
（４）次に３の温度条件に進む。ここではハイブリダイゼーションしなかったオリゴヌクレオチドを集めるコールドウォッシュ工程を行う。５６℃にてＢ＆Ｗ溶液５０μＬにオリゴヌクレオチドを抽出しＭＴＰ３５に出力した。これがｇｅｔ（Ｔ，−Ｓ）の出力オリゴヌクレオチド溶液である。
【０１７６】
（５）温度条件４のホットウォッシュ工程では４のようにさらに温度を上げて、磁気ビーズに捕獲されているオリゴヌクレオチドをコールドウォッシュと同様５０μＬのＢ＆Ｗ溶液に抽出してＭＴＰ３５に出力した。これがｇｅｔ（Ｔ，＋ｓ）の出力オリゴヌクレオチド溶液である。
【０１７７】
（ｃ）関数ｍｅｒｇｅ
ごく簡単なピペッティング操作で実行した。ピペッタで混合する溶液をそれぞれ吸引し、全て１箇所のＭＴＰのウェルに集めて混ぜることで実現した。
【０１７８】
（ｄ）関数ａｐｐｅｎｄ
図１６に示す反応である。ａｐｐｅｎｄするオリゴヌクレオチドが異なれば、別々の反応チューブで反応して同時に行った。
【０１７９】
（１）ａｐｐｅｎｄ反応する溶液を２０μＬ、ＭＴＰ３５よりピペッタで吸引し、サーマルサイクラ３８に吐出する。そのほか、ａｐｐｅｎｄ反応に必要な下記の溶液を反応用ウェル３８まで運搬した。
【０１８０】
ａｐｐｅｎｄ反応には、Ｎｅｗ Ｅｎｇｌａｎｄ Ｂｉｏ Ｌａｂｓ社のＴａｑリガーゼとその専用緩衝溶液とを用いた。
【０１８１】
反応溶液は以下の通りである。
【０１８２】
Ｔａｑリガーゼ（ＮＥＢ） ０．５μＬ（２０Ｕ）
反応ＤＮＡの溶液 原液２０μｌ
反応バッファ １２μＬ（１０倍希釈で使用、付属品）
連結オリゴヌクレオチド 各１０ｐｍｏｌ
ａｐｐｅｎｄオリゴヌクレオチド 各１０ｐｍｏｌ
滅菌蒸留水 全液量が１２０μＬになるように加えた。
【０１８３】
（２）ライゲーション反応を行った。サーマルサイクラの制御は下のように行った。ライゲーション反応を行ったのは温度条件２の時である。反応温度条件は以下の通りである。即ち、
１． ９５℃ １分
２． ５８℃ １５分（１から２までは１０℃／分で温度を下げる）
３． ２５℃ １０分（２から３までは１０℃／分で温度を下げる）
４． ７０℃ ３分（３から４までは１０℃／分で温度を上げる）
５． ７４℃ ３分（４から５までは１０℃／分で温度を上げる）
６． ８８℃ ３分（５から６までは１０℃／分で温度を上げる）
とした。
【０１８４】
（３）反応温度条件３の２５℃に下がってから磁気ビーズによる捕獲を行った。このとき、磁気ビーズを分散している溶液は捨て、ビーズをチップ内に保持したままサーマルサイクラ３８からライゲーション緩衝溶液を直接ピペッタで吸引した。すなわちライゲーション緩衝溶液中で捕獲反応を行った。緩衝溶液が捕獲のための液でないため念のため６０回吸引吐出を行い十分な量を捕獲できるようにした。磁気ビーズはサーマルサイクラ３８に保持した。
【０１８５】
（４）続いて１回目のコールドウォッシュを行った。核酸ハイブリッドの長さと溶液の塩濃度から７０℃が適温である。４の７０℃まで昇温したらサーマルサイクラ３８から磁気ビーズ液を吸引し、永久磁石でビーズを集め、溶液のみをＭＴＰ３５に吐出した。この溶液は廃液である。
【０１８６】
（５）そのまま、ピペッタでＭＴＰ３７のＢ＆Ｗ溶液５０μＬを吸引し、永久磁石を離して磁気ビーズを分散させてサーマルサイクラ３８に戻した。サーマルサイクラ３８では反応温度５にし、２回目のコールドウォッシュを実行した。ここでは（４）と同様に適温になったらサーマルサイクラ３８から吸引し、永久磁石でビーズを集めてＢ＆Ｗ溶液のみをＭＴＰ３５に吐出した。この溶液も廃液である。
【０１８７】
（６）このあと、さらに再びピペッタでＭＴＰ３７のＢ＆Ｗ溶液５０μＬを吸引し、永久磁石を離して磁気ビーズを分散させてサーマルサイクラ３８に戻した。サーマルサイクラ３８で反応温度６になったとき、溶液中にａｐｐｅｎｄ反応の済んだ１本鎖オリゴヌクレオチドが解離してくるのでこれをサーマルサイクラ３８から吸引し、永久磁石でビーズを集め、溶液のみをＭＴＰ２５に吐出し保持した。
【０１８８】
（ｅ）関数ｄｅｔｅｃｔ
図１８に示す反応と、キャピラリゲル電気泳動による検出でおこなった。最終的に得られた解を表すオリゴヌクレオチドを含むであろう溶液をテンプレートとし、ｇｒａｄｕａｔｅｄ ＰＣＲを行った。ＰＣＲはプライマーの組ごとに反応液を分けて行い、それぞれのＰＣＲ産物の有無および長さを検出することで解の配列を調べた。
【０１８９】
（１）解を含むと考えられるオリゴヌクレオチド溶液をテンプレートとする。プログラム中でａｍｐｌｉｆｙによりオリゴヌクレオチドの濃度は維持されているので、それからテンプレート濃度を推定し液量を定める。
【０１９０】
ＰＣＲ反応液の組成は、
ポリメラーゼ酵素（宝酒造） ０．５μＬ（２．５Ｕ）
増幅するＤＮＡの溶液 １ｆｍｏｌ程度の各々のオリゴヌクレオチドを含むだけの量
ｄＮＴＰ混合溶液 ８μＬ（２．５ｍＭ、付属品）
反応バッファ １０μＬ（１０倍希釈で使用、付属品）
プライマー 各オリゴヌクレオチドにｆｏｒｗａｒｄ，ｒｅｖｅｒｓｅ側とも５ｐｍｏｌずつ準備
滅菌蒸留水 全液量が１００μＬになるように加えた
とした。
【０１９１】
増幅には宝酒造のＰｙｒｏｂｅｓｔ ＤＮＡ Ｐｏｌｙｍｅｒａｓｅ ＰＣＲ増幅キットを用いた。反応温度条件は、
１． ９５℃ ３０秒
２． ５０℃ ３０秒
３． ７２℃ ６０秒 １〜３を３０サイクルであった。
【０１９２】
本実験で用いたプライマーは次の１２組である。
【０１９３】
（Ｘ_１ ^Ｔ，［Ｘ_２ ^Ｔ］）、（Ｘ_１ ^Ｔ，［Ｘ_２ ^Ｆ］）、（Ｘ_１ ^Ｔ，［Ｘ_３ ^Ｔ］）、
（Ｘ_１ ^Ｔ，［Ｘ_３ ^Ｆ］）、（Ｘ_１ ^Ｔ，［Ｘ_４ ^Ｔ］）、（Ｘ_１ ^Ｔ，［Ｘ_４ ^Ｆ］）、
（Ｘ_１ ^Ｆ，［Ｘ_２ ^Ｔ］）、（Ｘ_１ ^Ｆ，［Ｘ_２ ^Ｆ］）、（Ｘ_１ ^Ｆ，［Ｘ_３ ^Ｔ］）、
（Ｘ_１ ^Ｆ，［Ｘ_３ ^Ｆ］）、（Ｘ_１ ^Ｆ，［Ｘ_４ ^Ｔ］）、（Ｘ_１ ^Ｆ，［Ｘ_４ ^Ｆ］）。
【０１９４】
これらをＭＴＰ３５の異なるウェルに保持しておく。ＰＣＲ反応時にプライマー溶液はピペッタで吸引してサーマルサイクラ３８の異なるウェルに吐出し、反応に必要な溶液を加えてサーマルサイクラ３８を設定通りに動作させれば反応は完了する。この作業は容易に自動化できる。この後のキャピラリゲル電気泳動装置での試料のキャピラリヘの導入も自動化できる。キャピラリゲル電気泳動ではペックマン・コールター社のｄｓ１０００ゲルキットを使用した。プライマーＸ_１ ^ＴにはＦＩＴＣが標識されているため泳動像が観察された。今回の実施例では以上のｄｅｔｅｃｔの作業はキャピラリゲル電気泳動も自動では行わず、マニュアルで行った。以上の方法でプログラムの各関数を実行して得られた結果を図２１に示した。例えば関数ｄｅｔｅｃｔに対応する配列の同定方法としては、図１９に示すような手法がある。
【０１９５】
分子計算装置によるゲノム情報解析という計算パラダイムの有効性を示すために、実際に遺伝子の発現情報解析を行うＤＮＡ計算の実験を行った。計算はダイナミックプログラミングで３ＳＡＴの問題を解くときに使用するｇｅｔ、ａｐｐｅｎｄ、ａｍｐｌｉｆｙ、ｍｅｒｇｅおよびｄｅｔｅｃｔの基本命令を用いて実行することができる。最初の計算反応はエンコード反応である。単一試験管内で行える計算反応で、遺伝子の転写産物の情報がａｐｐｅｎｄ命令によりＤＣＮに変換される。変換テーブルは、アダプター分子Ａｉで表現され、２桁ｎ進数のＤＣＮへ１：１で変換される。２００種のＤＣＮは、２桁１００進数のＤＣＮとして利用することにより、最大１０，０００種類の遺伝子をエンコードすることができる。その後、ａｍｐｌｉｆｙ命令により増幅とｎ本の試験管への分注反応が行われる。最後にｎ本の試験管に対してａｐｐｅｎｄとｇｅｔ命令によりＤＣＮをデコードする計算反応が行われる。デコード反応はｎ本の試験に対して並列に行うことができる。移植断片対宿主病関連の転写産物に特異的な配列を有するＤＮＡオリゴヌクレオチドを入力データとして実験を行ったところ、計算反応が特異的且つ定量的に進むことが確認された。
【０１９６】
ＤＣＮに対するＤＮＡ計算により、遺伝子の発現情報解析を行う方法は、ＤＮＡチップで直接に転写産物分子を解析する方法に比べて優れた点をいくつか有している。性質が一様なＤＣＮに変換してから増幅するので、もとの頻度分布を崩さずに増幅して解析することが可能である。また、ＤＣＮデコードで「ｇｅｔ」命令を並列に行うためのＤＮＡチップは、同じＤＣＮの体系である限り同じものを使用することができる。その上、ＤＮＡプローブの数も大幅に少ない。ＤＮＡチップを作製する手間とコストが大幅に軽減される。更に、正規直交化されたプローブのハイブリダイゼーション反応は最適化されており、正確な計算処理を行うことが可能である。また、転写産物分子をラベルする必要はなく、その効率のばらつきによる誤差も生じない。これらの利点に加えて単なる発現プロファイルの計測にとどまらず、プロファイルに関する様々な情報解析をＤＣＮに対するＤＮＡ計算で行うことが可能である。
【０１９７】
なお、以上は５つの基本関数（関数ａｍｐｌｉｆｙ、関数ｇｅｔ、関数ｍｅｒｇｅ、関数ａｐｐｅｎｄ、関数ｄｅｔｅｃｔ）を例に挙げたが、これらに限定されるものではない。例えば、関数ｅｎｃｏｄｅ、関数ｄｅｃｏｄｅ、関数ｃｌｅａｖｅなども、分子計算機における計算を定義するのに有効である。
【０１９８】
関数ｅｎｃｏｄｅとは、分子の入った試験管Ｔに特定の配列ｓ_ｉを含む分子が存在したときに、それに１：１に対応した配列ｃ_ｉを出力する関数であり、例えばｅｎｃｏｄｅ（Ｔ，ｓ_１，ｓ_２，…，ｓ_ｎ，ｃ_１，ｃ_２，…，ｃ_ｎ）などで表現される。
【０１９９】
関数ｄｅｃｏｄｅとは、分子の入った試験管Ｔに特定の配列ｓｉを含む分子が存在したときに、ｓｉに隣接する配列ｃｉの相補配列を含む分子を出力する関数であり、例えばｄｅｃｏｄｅ（Ｔ，ｓｉ）などで表現される。
【０２００】
関数ｃｌｅａｖｅとは、分子の入った試験管Ｔの中の特定の配列ｓを含む分子を切断する関数であり、例えばｃｌｅａｖｅ（Ｔ，ｓ）などで表現される。この関数ｃｌｅａｖｅは２通り存在する。第１は、試験管Ｔの中の分子を特定の配列ｓで切断する関数、第２は、分子の入った試験管Ｔの中の特定の配列から特定数の塩基だけ離れた配列を切断する関数である。この関数ｃｌｅａｖｅは、分子計算部２２では、制限酵素の切断として定義される。
【０２０１】
もちろん、この３つの基本関数を含めた８つの基本関数のみで表現する必要は無く、他の基本関数を設定することも可能である。
【０２０２】
以下の表２に、基本関数と、電子計算部２１における処理体系としての表現、分子計算部２２における処理体系としての表現を示す。なお、分子計算部２２における処理体系は、表２に示したものに限定されるものではなく、あくまで電子計算部２１における処理体系を表現する一態様である。
【０２０３】
表２に示すように、基本関数（手続または関数）は、符号分子の演算反応に対応付けて定義され、変数及び定数は、分子を核酸配列に基づき符号化した符号分子に対応付けて定義されている。
【０２０４】
【表２】

【０２０５】
このような本発明の態様によれば、上述したような分子計算機の高い並列計算性を活かし、しかも分子計算機だけでは実現することが難しい機能を電子計算機により補完することにより、操作する者が分子計算のための反応操作手順や符号化分子の割り当て等を行う必要はなくなる。
【０２０６】
また、本計算装置を用いて遺伝子解析を行った場合、特定の配列を有する核酸を標的としてそれらの存在または不存在を評価することや、またそれによって、例えば、遺伝子型や遺伝子の発現の状態を決定することが、小さい実験誤差で、低コストで、且つ簡便に行うことが可能である。
【０２０７】
更にまた、本発明は、上述の記載に基づいて以下の方法および分子計算用ソフトウェアも提供する。即ち、電子計算部と分子計算部とを、電気学的プログラムが認識可能に表現された分子情報に基づいて一体的に機能させることを特徴とする分子計算方法を提供するものである。
【０２０８】
また、電子計算部と分子計算部とを具備する分子計算装置に適用するためのソフトウェアであって、前記電子計算部および／または前記分子計算部に適用され、前記電子計算部による計算作業と前記分子計算部による計算作業とを、各計算部で電気学的に認識可能な情報形式で機能させることを特徴とする分子計算用ソフトウェア、詳しくは、分子計算部により計算した情報を、電子計算部の電気学的プログラムに適合するような情報形式に変換する機能を有することを特徴とする分子計算用ソフトウエア、更に詳しくは、電子計算部により計算した情報を、分子計算部の計算作業に適合するような情報形式に変換する機能を有する分子計算用ソフトウエアを提供することである。
【０２０９】
本発明の分子計算ソフトウェアを用いることにより、上述の本発明の計算装置の実行を容易に行うことが可能である。また、当該分子計算ソフトウェアは、本発明の計算装置を総括して管理しても、またその構成要素の一部分を独立して管理しても、または構成要素の幾つかの部分を組み合わせて連動して管理してもよい。
【０２１０】
ＩＩ．遺伝子解析の適用例
１．概要
本発明者らは、遺伝子の解析を、核酸分子を入力データとして用いる計算として捉えるというオリジナルなアイデアを着想した。この着想は、Ｉで示した分子計算装置を用いて実現可能である。例えば、発現ｍＲＮＡおよびゲノム上の遺伝子を符号核酸のデータに変換し、それらデータの論理和、論理積、否定を計算することで、遺伝子型判定や特定の疾患での遺伝子の発現条件を求めることができる。
【０２１１】
以下に示す２．〜４．の第１の例〜第３の例は、Ｉに示した分子計算装置を適用して実現される。
【０２１２】
２．分子計算装置を遺伝子解析に適用する第１の例
本発明の１態様に従う分子計算装置に適した問題の例は、ＮＰ完全問題や３ＳＡＴ等の純粋に数学的問題、ゲノム情報解析のように入力データが核酸分子で与えれられるような問題、機能性分子の設計、および機能の評価を電子コンピュータで行うことが困難な問題等である。
【０２１３】
以下に、分子計算装置により行う更なるゲノム情報解析の例を示す。まず、ゲノム情報を正規直交化ＤＮＡ塩基配列で表現した数字である体系に変換する。その後、そのＤＣＮに対するＤＮＡ計算を純粋に数学的問題を解くときのように行い、その計算結果からものゲノム情報を解析する。
【０２１４】
この方法では、以下に示すような２種類の核酸プローブ、即ち、プローブＡおよびプローブＢが使用される（図２２（ａ））。
【０２１５】
プローブＡは、標的核酸の一部の領域の塩基配列Ｆに相補的な配列Ｆ’と、これに結合した結合分子からなる。
【０２１６】
ここで、結合分子は、互いに特異的に高親和性を有する２つの物質の内の何れか一方の物質である。例えば、ビオチンまたはアビジン若しくはストレプトアビジン等である。また、結合分子は、直接に配列Ｆ’に結合しても、或いは任意の配列を解して間接的に配列Ｆ’に結合してもよい。間接的に結合する場合の任意の配列は、如何なる塩基配列であっても、如何なる塩基数であってもよい。好ましくは、標的核酸上の塩基配列に非相補的な配列である。
【０２１７】
プローブＢは、標的核酸の一部の領域の塩基配列Ｓに相補的な配列Ｓ’とフラッグとからなる。本例におけるフラッグは二本鎖からなる。前記二本鎖は、複数のユニットからなる任意の配列を有す。また、フラッグは、これ自身が標的核酸と結合はせず、また、これらは互いに如何なる相互作用も示さないことが必要である。
【０２１８】
本方法に使用される配列Ｆ’および配列Ｓ’は、１以上の塩基数を有し、より好ましくは１５以上の塩基数を有する。
【０２１９】
ユニットの設計例を図２３（ａ）に示す。フラッグＦＬの複数のユニットの各１ユニットは、１０塩基数以上としてよく、より好ましくは約１５塩基数である。フラッグＦＬのユニット数は、何れでもよいが、解析の容易さから４ユニットが好ましい。しかし、これに限られるものではない。
【０２２０】
複数の標的核酸を同時に検出する場合には、多種類のユニットを組み合わせてフラッグＦＬを構築する。たとえば、ＳＤ、Ｄ０、Ｄ１、ＥＤの４ユニットからなるフラッグＦＬを設計する場合を例とすると、先ず、２２種類のユニットを設計し、その中から２種類を選択してプライマーとなるＳＤユニットと、もう１つのプライマーであるＥＤユニットとする。残りの２０種類のユニットを用いて、各標的核酸の種類毎に、選択する２つのユニットの種類を変えることによりＤ０、Ｄ１を設計すると、１００種類の異なる核酸配列を検出することが可能である（図２３（ａ））。
【０２２１】
２２種類のユニットは、正規直交化された塩基配列により設計することが好ましい。正規直交化された塩基配列はＴｍ値が揃っており、相補配列以外とは安定したハイブリッドは形成しない。また、相補配列とのハイブリッド形成を阻害するような安定した２次構造は形成しない。これにより、最終的な検出時のミスハイブリを少なくし、ハイブリの形成速度を上げることが可能になる。したがって、検出精度を向上すること、および検出時間を短縮することが可能になる。また、ユニット数を増すことや、ユニットの種類を増すことにより、１００００種類の異なる核酸配列をも検出することが可能である。
【０２２２】
図２２（ａ）〜図２２（ｉ）を用いて、本例の方法を更に説明する。図２２（ａ）に、４ユニットからなるフラッグＦＬの例を示した。該４ユニットは、ポリメラーゼ連鎖反応（ｐｏｌｙｍｅｒａｓｅ ｃｈａｉｎ ｒｅａｃｔｉｏｎ；以後、ＰＣＲ増幅またはＰＣＲ反応と称す）においてプライマーとなるＳＤユニットと、標的核酸の種類を認識するための認識用ユニット、即ち、Ｄ０ユニットおよびＤ１ユニットと、およびもう１つのプライマー配列であるＥＤユニットとからなる。これらの各ユニットは、後の工程においては、夫々が読み取り枠となる。
【０２２３】
検出は、まず上記のプローブＡとプローブＢを、標的核酸と混合する（図２２（ａ））ことにより行う。このとき、試料に含まれる標的核酸は、複数の異なる核酸分子群であってもよい。例えば、検出されるべき標的核酸の種類が１００種類以下であるならば、Ｄ０ユニットは、Ｄ０−１からＤ０−１０の１０種類の中から選択され、且つＤ１ユニットは、Ｄ１−１からＤ１−１０の１０種類の中から選択される（図２３（ａ））。
【０２２４】
次に、プローブＡ、プローブＢ、および標的核酸をハイブリダイゼーションに適した条件で一定時間インキュベーションし、ハイブリダイゼーションを行なう（図２２（ｂ））。ハイブリダイゼーションの条件は、例１に示す通りでよい。
【０２２５】
かかるハイブリダイゼーションにより、プローブＡおよびプローブＢの両方が同一の標的核酸上に結合する（図２２（ｂ））。
【０２２６】
次に、標的核酸にハイブリダイズしたプローブＡおよびプローブＢを連結する（図２２（ｃ））。連結の条件は、例１に示す通りでよい。
【０２２７】
また、フラッグＦＬのＴｍ値は、配列Ｆ’およびＳ’より高い温度に設計することが好ましい。これにより本検出方法におけるハイブリダイゼーション、ライゲーションおよび変性等の操作の加熱または冷却の際に、検出感度の低下をもたらすフラッグの変性を防ぐことが可能である。
【０２２８】
次に、得られたフラッグＦＬの情報をＢ／Ｆ分離する（図２２（ｄ））。具体的には、プローブ（Ａ＋Ｂ）に具備される結合分子を、その対となるべき結合分子を介して固相担体に補足する（図２２（ｅ））。
【０２２９】
前記固相担体は、基板、ビーズ等の粒子、容器、繊維、管、フィルター、アフィニティ・カラム、電極等を用いることが可能であるが、好ましくはビーズである。
【０２３０】
次に、結合分子に捕捉された状態で、プローブ（Ａ＋Ｂ）のフラッグＦＬを変性し一本鎖にする（図２２（ｆ））。得られた液相中の一本鎖配列ＦＬ’に対してＰＣＲ増幅を行なう（図２２（ｇ））。上述したように、予めフラッグＦＬには、２つのプライマー配列ＳＤおよびＥＤが配置してある。従って、このプライマー配列を利用してＰＣＲ反応が容易に行ない得る。また、このとき、ＰＣＲに使用する２つのプライマーの一方、たとえばＳＤ配列に、ビオチン等の結合分子を結合しておくことが好ましい。このときのＰＣＲの詳細な条件は、設計したフラッグＦＬに依存する。
【０２３１】
続いて、該ＰＣＲ反応の終了後、結合分子を固相した固相担体に結合することによって、ＰＣＲ産物である二本鎖配列を回収する（図２２（ｈ））。ここで、固相化された担体は、前記結合分子と対になる結合対のもう一方の物質である。さらに、変性により配列ＦＬ’を除き、一本鎖配列ＦＬのみを固相担体上で回収する（図２２（ｉ））。
【０２３２】
続いて、固相上の一本鎖フラッグ配列ＦＬの解析を行なう。まず、一本鎖フラッグ配列ＦＬが結合した前記固相担体を１０等分する（Ｄ１ユニットがＤ１−１からＤ１−Ｄ１０の場合）。各々に、標識分子と結合したＤ１−１’からＤ１−１０’配列の一つおよび全てのＤ０’配列（Ｄ０−１’からＤ０−１０’）を加え、フラッグ配列ＦＬにハイブリダイズする。
【０２３３】
続いて、ハイブリダイズした２つの核酸分子をライゲーションにより連結する。ここで、ライゲーションの条件および標識物質に関する定義は上述した通りである。その後、変性により連結された分子を液相に回収する。
【０２３４】
得られた標識された核酸分子の解析は、予めＤ０−１からＤ０−１０の核酸分子を固相化したＤＮＡチップまたはＤＮＡキャピラリ等に対して、ハイブリダイズすることにより行うことができる。特に、ＤＮＡキャピラリは、Ｄ０−１からＤ０−１０で１０等分に分けられたものを同時に処理できるので、これにより分析は容易になるであろう。
【０２３５】
例えば、各１０種類のＤ０−１からＤ０−１０と、Ｄ１−０からＤ１−１０の配列を用いてフラッグＦＬを設計した場合、図２３（ａ）の１の位置にはＤ０−１に相当する配列が固定され、標識されたＤ１−１’分子と連結された拡散分子６３にハイブリダイズされる。同様に、他の位置には列により相当するＤ０配列が固定され、行により相当するＤ１’分子と連結された拡散分子にハイブリダイズされる。このような行列の配置を、後述するＤＮＡキャピラリに対して用いる（図２３（ｂ））と解析が容易に行える。
【０２３６】
ここでは、１０種類のユニットを用いた例を挙げたが、ユニットの種類は１０種類に限られるものではなく、それ以下でも、それ以上でもよい。
【０２３７】
ここで使用する「ＤＮＡキャピラリ」とは、標的核酸を検出するための装置であり、その内側に該標的核酸に対する相補的配列が結合されており、該相補的配列に標的核酸を結合することにより、該標的分子を検出する装置をいう。図２３（ｂ）に示す通り、多数のＤＮＡキャピラリを同時に使用し、且つ斜線で示した部分に、互いに異なるプローブを配置することにより、同時に多くの標的核酸を検出することが可能である。
【０２３８】
また、本方法では、フラッグ配列ＦＬの各ユニットには正規直交化配列が使用されているので、実施されるハイブリダイズの反応温度等の条件を均一化することが可能である。これにより、ミスハイブリを防止でき、高い精度が得られる。また、同一条件の下で一度に多くの解析を行なうことが可能であるため、検出時間の短縮化を達成することが可能である。また、本方法により複雑なゲノム情報をＤＮＡの塩基配列で表現した数値に変換することも可能となり、ＤＮＡ分子反応を利用した計算を行うことにより、多種類の情報や、互いに連鎖した複雑な遺伝子情報を容易に解析することが可能になる。また、コード化したのちに容易にコード化核酸を増幅できるので、少ないコピー数の標的配列であっても正確に且つ定量的に検出することが可能である。また、コード化することにより、多くの情報を圧縮することが可能である。従ってＤＮＡチップまたはキャピラリーアレイ等の検出手段の所要数を節約することが可能である。
【０２３９】
ここで使用する「エンコード反応」とは、ある塩基配列を、正規直交化塩基配列で表現されるコードに変換することをいう。上述の図２２（ａ）〜図２２（ｆ）の工程がこれに相当する。
【０２４０】
また、ここで使用する「デコード反応」とは、前記で変換されたコードの読み取りを行ない、それにより元の情報を復元することをいう。上述の図２２（ａ）〜図２２（ｉ）の工程がこれに相当する。
【０２４１】
この方法では、上述したような１種類の標的核酸を検出するのみに留まらず、複数種類のフラッグ配列を設計すれば、同様な工程を経ることにより複数種類の標的核酸を同時に検出することも可能である。
【０２４２】
３．分子計算装置を遺伝子解析に適用する第２の例
本発明の更なる１側面では、上述のような分子計算装置を用いて、ゲノム情報解析を行う一般的な計算方法論が提案される。特に、該ゲノム情報解析方法では、以下のような利点が得られる。即ち、そのような方法では、先ず、特定の遺伝子の塩基配列に対して、任意に設計した所望配列を任意に割り当てる。そして、その割り当てに従って、該特定の遺伝子の塩基配列を設計された配列に変換し、変換後に得られた安定性の高い配列を演算に使用することが可能である。従って、反応の設計における自由度が高くなり、且つ正確な反応を実施することが可能になる。
【０２４３】
（ａ）第１の実施の形態
（１）概要
遺伝子解析に適用する本発明の第１の実施の形態について述べる。第１の実施の形態は、核酸分子による演算によって、遺伝子の有無を判定する遺伝子解析の例を示す。
【０２４４】
その概要は以下の通りである。まず、細胞で発現された遺伝子群を基にｃＤＮＡ群を作製する。得られたｃＤＮＡ群に含まれる発現遺伝子と、含まれない非発現遺伝子に関する情報、即ち、標的遺伝子の有無に関する情報を、人工的に設計した配列をもつＤＮＡ分子の形態に変換し、表現様式を変更する。この変換によって得られたＤＮＡ分子を、演算用核酸に対してハイブリダイゼーションする。以上の過程が演算解析の過程である。ここで、前記ＤＮＡ分子は、特定の標的遺伝子が存在しているか否かの情報を担う一種の信号として機能する。
【０２４５】
例えば、ある標的遺伝子が存在することを確認することによって、当該遺伝子が発現遺伝子であることが判定できる。或いは、その標的遺伝子が存在しないことを確認することによって、当該遺伝子が非発現遺伝子であることが判定できる。従って、本解析方法では、サンプル中に含まれる標的分子を検出することが可能であるばかりではなく、同時に、サンプル中に含まれない標的分子に関しては、それが存在しないとい情報を得ることが可能である。
【０２４６】
（１．１）準備
本発明の１態様である計算方法には以下のような分子が必要である。実質的な計算に先駆けて、以下の分子を調製することが必要である。当該調製はそれ自身公知の方法により行うことが可能である。
【０２４７】
溶液に含まれるｃＤＮＡを検出するために図２４に示す２つのプローブ、即ち、点線で囲まれたａ_ｉとＡ_ｉを準備する。ａ_ｉは、標的のｃＤＮＡの一部の配列に相補的な配列を含み且つ５’端にビオチンを標識したオリゴヌクレオチドである。Ａ_ｉなるオリゴヌクレオチドは部分的にハイブリダイゼーションにより２本鎖になったオリゴヌクレオチドである。Ａ_ｉを構成する２本鎖のうちの一方のオリゴヌクレオチドは、人工的に設計されたＳＤ、ＤＣＮ_ｉおよびＥＤなる塩基配列を３’端側に有し、標的のｃＤＮＡの一部分の配列に相補的であり且つａ_ｉ分子の標的に相補的な配列に隣接するような配列を５’端側に有する。また、前記人工的に設計した塩基配列は、当該相補的な配列よりも３’末端側に配置される。また、Ａ_ｉの標的ｃＤＮＡに相補的な配列の５’端はリン酸化されている。２本鎖Ａ_ｉを構成するもう１方の鎖はＳＤ、ＤＣＮ_ｉおよびＥＤの配列に相補的な配列をもつオリゴヌクレオチドである。ａ_ｉおよびＡ_ｉは、検出したい標的遺伝子毎に任意に設計する。ここでいう「標的遺伝子」とは、溶液中に存在するまたは存在しないことを検出したい遺伝子である。またこのとき、ＤＣＮ_ｉの配列は標的ごとに異なる配列になるように設計し、ＳＤおよびＥＤはすべてのＡ_ｉで共通する配列になるように設計する。これらの人工的な配列は、任意に設計可能であるので、所望するＴｍ値を設定することが可能である。従って、安定に且つミスハイブリダイゼーションの少ない反応を行うことが可能である。
【０２４８】
更に、図２８に示すような５’端にビオチン標識をしたＳＤ配列と同じ配列を有するプライマー１と、ＥＤ配列に相補的な配列を有するプライマー２が必要である（図２８）。
【０２４９】
また更に、図３１に示すような「標的が存在すること」を示すＤＣＮ_ｉに相補的な配列を有する存在オリゴヌクレオチド３（図３１）と、「標的が存在しないこと」を示すＤＣＮ_ｉ ^＊に相補的な配列を有する不存在オリゴヌクレオチド６（図３５）が必要である。
【０２５０】
また、図３２に示すような反転オリゴヌクレオチド４が必要である。これは、ＤＣＮ_ｉに対応するように人工的に設計され且つＤＣＮ_ｉとは異なる配列を有した塩基配列であるＤＣＮ_ｉ ^＊を５’端側に具備し、その３’端側にはＤＣＮ_ｉ配列を具備するオリゴヌクレオチド（図３２）である。
【０２５１】
（１．２）存在分子と不存在分子への変換
本発明の方法では、まず、サンプル中に、特定の標的分子が存在しているという情報を「存在分子」に変換し、特定の標的分子がある系に存在していないという情報を「不存在分子」に変換する。ここで使用する「存在分子」と「存在オリゴヌクレオチド」の語は互いに交換可能に使用される。また「不存在分子」と「不存在オリゴヌクレオチド」も同様に交換可能に使用される。
【０２５２】
このような分子の存在、不存在を分子形態表現に変換する方法について図２４から図３５を用いて説明する。図２４から図４５は、それぞれの工程における系に存在する分子を模式的に示したものである。
【０２５３】
図中、ＤＮＡを矢印により示し、矢印の元部をＤＮＡの５’端とし先端を３’端とする。矢印の途中に入る該矢印への短い垂線は、塩基配列の区切りを示す部分である。また、図中の矢印の近くに示す、「ａ」、「Ａ」および「ＤＣＮ」等のアルファベットは配列の名前を示す。また、「ａ」、「Ａ」および「ＤＣＮ」等のアルファベットに付された添え字「ｉ」および「ｋ」は整数であり、それぞれの配列が、どの遺伝子に対応するかを表示するために付されたものである。ここでは「ｉ」および「ｋ」により任意の配列が示される。また、ここでは便宜的に「ｉ」は発現遺伝子を、「ｋ」は非発現遺伝子を示す。また、図中、配列名の上に線が引かれている場合は、相補的な配列を示す。図中の斜線のある円はビオチン分子を示し、白い大きい円は磁気ビーズを現す。磁気ビーズから右横に伸びる黒い十字は、該磁気ビーズに固定されたビオチン分子と特異的に結合するストレプトアビジン分子を模式的に示している。
【０２５４】
ａ．「標的が存在する」という情報の「存在分子」への変換
標的が存在する場合の存在分子への変換は図２４から図４２に示す工程を逐次的に行うことにより実施される。
【０２５５】
まず、図２４を参照されたい。上述の通り合成したａ_ｉおよびＡ_ｉを、Ｔａｑライゲースのような高温で活性の高い酵素の反応バッファ中でｃＤＮＡと反応させる。但し、このライゲーション反応の温度はＡ_ｉオリゴヌクレオチドの２本鎖部分が解離しない温度とする。この反応の結果、標的が存在した場合は、図２５のようにライゲースによりａ_ｉとＡ_ｉは連結される。次に、この反応溶液から図２６に示すようにストレプトアビジンを表面に結合した磁気ビーズにて前記連結オリゴヌクレオチドを抽出する。このとき、未反応のａ_ｉ分子もビーズに捕獲されるが、以後の反応には関係しない。
【０２５６】
続いて、熱をかけることで、磁気ビーズで捕獲したＡ_ｉとａ_ｉの連結分子からＡ_ｉ部分の相補鎖を分離抽出する（図２７）。この操作によって、最初の溶液にｃＤＮＡが存在していれば、それに対応するＤＣＮ_ｉ配列に相補的な配列を含んだオリゴヌクレオチドが抽出される（図２７）。この抽出オリゴヌクレオチドをテンプレートとして、５’端にビオチン標識をしたＳＤ配列と同じプライマーと、ＥＤ配列に相補的なプライマーにて図２８のようにＰＣＲ増幅反応を行う（図２８）。これにより存在していると判明した遺伝子を検出するＤＣＮ_ｉ配列が増幅される。
【０２５７】
このＰＣＲ増幅による２本鎖の産物を、図２９に示すようにストレプトアビジン結合磁気ビーズにより捕獲する（図２９）。捕獲した２本鎖のＰＣＲ産物を捕獲したままで熱をかけて１本鎖にし、解離させた相補鎖を緩衝液交換により除去する（図３０）。続いて図３１のように、ＤＣＮ_ｉに相補的な配列をもつ存在オリゴヌクレオチドを、ビーズに捕獲されたＰＣＲ産物にハイブリダイズする（図３１）。このハイブリダイゼーションの後、過剰な存在オリゴヌクレオチドを除去し、続いて、改めて熱をかけることによってビーズに捕獲されているＤＣＮ_ｉの相補鎖（即ち、存在オリゴヌクレオチド３）をバッファ中に抽出する。ここで抽出されたＤＣＮ_ｉに相補的な塩基配列を有する存在オリゴヌクレオチド３が、もとのｃＤＮＡ溶液中に標的遺伝子が存在することを示す存在分子である。
【０２５８】
ｂ．「標的が存在しない」という情報の「不存在分子」への変換
上述の工程により、存在した標的遺伝子を、それが存在するという情報を示す存在分子（即ち、存在オリゴヌクレオチド）に変換した後で、存在しないという情報をこれを示す分子（即ち、不存在オリゴヌクレオチド）に変換する。標的が存在しない場合には、存在しなかったことを示す不存在オリゴヌクレオチドが抽出される。この抽出は以下の通りに実施される。
【０２５９】
予め、図３２にあるような反転オリゴヌクレオチドを全ての検出対象の遺伝子のＤＣＮについて準備する。上述した通り、反転オリゴヌクレオチドは、ＤＣＮ_ｉに対応するように人工的に設計され、且つＤＣＮ_ｉとは異なる配列を有した塩基配列ＤＣＮ_ｉ ^＊を５’端側に具備し且つ３’端側に隣接してＤＣＮ_ｉ配列を具えたオリゴヌクレオチドである。このような反転オリゴヌクレオチドと存在分子とのハイブリダイゼーション反応を利用することにより、「存在しない標的」を検出可能な「不在分子」に変換することが可能である。
【０２６０】
まず、図３２に示す工程において、反転オリゴヌクレオチド４に対して、図３１の過程で抽出した発現遺伝子に対応するＤＣＮ_ｉの存在オリゴヌクレオチド３３をハイブリダイズし、ポリメラーゼにより伸長反応を行う（図３２）。その結果、発現遺伝子のＤＣＮ_ｉは伸長し、ＤＣＮ_ｉ ^＊の配列の部分まで相補鎖が合成される（図３２）。一方、図３３の通り、標的分子が非発現遺伝子（ここではＤＣＮ_ｋと示す）であった場合、ＤＣＮ_ｋに相補的なオリゴヌクレオチドは反応液中に存在しないため、反転オリゴヌクレオチド５は１本鎖のままで存在する（図３３）。これら２本鎖と１本鎖の混合物は、ヒドロキシアパタイトを含むカラムに通すことで１本鎖の反転オリゴヌクレオチド５のみを抽出することができる（図３４）。
【０２６１】
このように抽出した非発現遺伝子に対応するＤＣＮ_ｋをもつ反転オリゴヌクレオチド５を、ストレプトアビジンを結合した磁気ビーズに捕獲する（図３５）。次に、上述した存在オリゴヌクレオチド３のみの抽出と同様に、ＤＣＮ_ｋ ^＊に相補的なオリゴヌクレオチド６をハイブリダイゼーションし、過剰なオリゴヌクレオチドを除去して非発現遺伝子を示すＤＣＮ_ｋ ^＊にハイブリダイゼーションした不存在オリゴヌクレオチド６のみを抽出することができる（図３５）。
【０２６２】
不存在オリゴヌクレオチド６を得るための工程は以下のようにも実施できる。即ち、ＤＣＮ_ｉオリゴヌクレオチドの５’端にＦＩＴＣなどの蛍光分子を標識しておき、ＤＣＮ_ｉに相補的な配列を有するプローブを含むＤＮＡマイクロアレイにおいてハイブリダイゼーション反応を行う。これをスキャナなどで読み取り、どのＤＣＮ_ｉが存在するかを検出する。このとき同時に、これにより存在していないＤＣＮ_ｋもわかる。従って、これらのデータから、次の演算のためにＤＣＮ_ｋ ^＊なる不存在オリゴヌクレオチド６を準備する。以上により存在しない核酸をその核酸に対応する不存在を示す核酸に変換することが達成される。これにより演算用核酸上での論理演算が可能になる。
【０２６３】
また、この不存在オリゴヌクレオチド６を作る工程において、１本鎖の反転オリゴヌクレオチド５を鋳型として不存在オリゴヌクレオチド６を増幅してもよい。増幅は、例えば、図４０から図４５に示される各工程を経て実施することが可能である。ここで、図４０から図４５は、それぞれの工程を示すものであり、且つ各系に存在する分子を模式的に示したものである。各図中に示される記号等の詳細は上述した通りである。まず、図３４の工程に従って、抽出された１本鎖のままの反転オリゴヌクレオチド５を得る（図３４）。この反転オリゴヌクレオチド５に対して、図４０に示すような、３’端側でＳＤに相補的な配列と、且つ５’端側でＥＤ配列に相補的な配列と結合したＤＣＮ_ｋ ^＊に相補的な配列を有したオリゴヌクレオチド７をハイブリダイゼーションすることで不存在オリゴヌクレオチド６を抽出する。続いて、ＦＩＧ．３６から２２に示すような工程により、存在オリゴヌクレオチド３と同様に増幅したＤＣＮ_ｋ ^＊を得ることができる。即ち、図４１に示す工程で、図４０の工程において得たオリゴヌクレオチド７に対してビオチン標識したＳＤ配列を有するプライマー１と、ＥＤ配列に相補的な配列を有するプライマー２を用いてＰＣＲ増幅する（図４１）。次に、ＰＣＲ産物をビオチンをストレプトアビジン分子に結合することにより回収する（図４２）。続いて、熱変性により、ＰＣＲ産物を１本鎖にする（図４３）。続いて、ＤＣＮ_ｋ ^＊に相補的な配列をもつ不存在オリゴヌクレオチドを、ビーズに捕獲されたＰＣＲ産物にハイブリダイズする（図４４）。このハイブリダイゼーションの後、過剰な存在オリゴヌクレオチドを除去し、続いて、改めて熱をかけることによってビーズに捕獲されているＤＣＮ_ｋ ^＊の相補鎖（即ち、不存在オリゴヌクレオチド６）をバッファ中に抽出する。ここで抽出されたＤＣＮ_ｋ ^＊に相補的な塩基配列を有する不存在オリゴヌクレオチド６が、即ち、もとのｃＤＮＡ溶液中に標的遺伝子が存在しないことを示す不存在分子である。
【０２６４】
（１．３）演算工程
演算工程では、上述で得られた存在分子および不存在分子と、以下に説明する演算用核酸とのハイブリダイゼーションおよび相補鎖合成とを行い、それによって、所望する条件を現す演算式を解き、該条件を満たす解を求める並列計算を行う。
【０２６５】
演算の例として演算式として数式４を用いる。ある特定の配列ＤＣＮ_１、ＤＣＮ_２、ＤＣＮ_３およびＤＣＮ_４を標的配列とし、その有無の条件について論理式として示した数式４を、演算用核酸と存在分子および不存在分子とのハイブリダイゼーション反応と伸長によって、演算し、値を評価する。数式４は、ＤＣＮ_１、ＤＣＮ_２、ＤＣＮ_３およびＤＣＮ_４についての所望する有無についての組合せを所望の条件として示している。即ち、本例では数式４の解を求めるということは、あるサンプル中における複数の標的配列の有無を同時に評価することである。
【０２６６】
【数４】

【０２６７】
式中、「¬」は「否定」、「∧」は「論理積」、「∨」は「論理和」を表す記号である。
【０２６８】
演算用核酸に設定した条件を満たす場合は、当該式の値は「１」、即ち「真」となる。また、演算用核酸に設定した条件が満たされない場合は、当該式の値は「０」、即ち「偽」となる。また本発明は、基本的な排他的論理和を達成している。従って、この組合せを用いればブール代数の全ての論理演算を実現できる。
【０２６９】
図３６に示すのが演算用核酸８の配列構造である。演算用核酸は１本鎖のオリゴヌクレオチドであり、図では矢印で示している。矢印の向きは５’端から３’端に向かっている。５’端にはビオチン分子が付いている。演算用核酸は複数のユニットを含む。その塩基配列は、５’端から順に、マーカー分子が結合するＭ_１、ＤＣＮ_１が存在しないときに得られるオリゴヌクレオチドを検出する配列ＤＣＮ_１ ^＊、ＤＣＮ_２、ポリメラーゼによる相補鎖伸長が止まるような配列を具えたストッパ配列Ｓ、２つ目のマーカーが付くＭ_２、ＤＣＮ_３、ＤＣＮ_４ ^＊である。この演算用核酸の配列は、論理式の各項の並びにほぼ対応する。論理式の「否定」もそのままの配列となる。例えば、「ＤＣＮ_４」配列の存在が「否定」される場合には、「ＤＣＮ_４ ^＊」の配列を用いる。ここで、ＤＣＮ_４ ^＊は上述した通りのＤＣＮ_４の対応するように設計された人工的な配列である。また、「論理和」記号はＳ配列に置き換えればよい。「論理積」は演算用核酸上の配列として置き換える必要はない。式１のための演算用核酸の演算は、表３の各条件を満たすときに、その値が１になる。表３中「―」はどのような状態でもよいことを示す。
【０２７０】
【表３】

【０２７１】
以下、論理式を評価する核酸反応について、その工程を説明する。演算工程は図３７から図３９の工程を含み、演算反応は次の手順で行う。先に述べた論理式の演算を行うときには、図３６に挙げた配列を具備した演算用核酸を準備する。１つのチューブに対して１種類の演算用核酸を入れ、それに先の工程で得た存在オリゴヌクレオチドと不存在オリゴヌクレオチドを含む溶液を入れ、ハイブリダイゼーション反応を行う（図３７）。ここで、仮に、ＤＣＮ_１、ＤＣＮ_３、ＤＣＮ_４が存在し、ＤＣＮ_２が不存在であるならば、演算用核酸にハイブリダイゼーションするのはＤＣＮ_３のみである（図３７）。また、ハイブリダイゼーション反応後、Ｔａｑポリメラーゼなど高温でも活性のある酵素により、ミスハイブリダイゼーションがおきない条件下で伸長反応を行うと、図３８のようにＭ_２配列部分は２本鎖となりＳ配列で伸長が止まる（図３８）。
【０２７２】
次に、ストレプトアビジン磁気ビーズにより反応が終了した演算用核酸を捕獲する（図３９）。最後に、検出反応としてマーカーオリゴヌクレオチドのハイブリダイゼーション反応を行う（図３９）。図３９では、担体上に固定された演算用核酸を示しており、その５’端のビオチン分子は担体上にあるストレプトアビジン分子に結合している。さらにマーカー検出配列に相補的な配列をもつマーカーオリゴヌクレオチドＭ_１、Ｍ_２を準備する。これらマーカーオリゴヌクレオチドの５’端には蛍光を発する分子が付いている。この例では、演算用核酸のＭ_１配列は２本鎖化されていないので、当該マーカーは演算用核酸に結合することが可能である（図３９）。このあと、結合していないマーカーを除去してビーズを蛍光観察すれば、マーカーオリゴヌクレオチドＭ_１がハイブリダイゼーションして蛍光を発するため演算結果得られる論理式の値は「１」であることが分かる。
【０２７３】
上述では、Ｓ配列をストッパーとして使用したが、Ｓ配列を必ずしも配置する必要はない。その代わりとしてＳ配列をなくし、相補鎖が形成されないように人工的な塩基を具えたヌクレオチドを含ませてもよい。このとき、演算用核酸をＳ配列の両側に配置すればよい。例えば、Ｓ配列は、他の配列部分にシトシン塩基が含まれないように設計し、Ｓ配列の塩基配列にシトシン塩基が含まれるように設計してもよい。この場合、図３８に示す演算用核酸上での伸長反応の際にモノマーとしてｄＧＴＰを別途加えなければ伸長反応はＳ配列上で停止する。また或いは、ポリメラーゼが伸長反応を停止し易いグアニンやシトシンの連続した塩基配列としてもストッパーとしての目的を達成できる。或いは、Ｓ配列に相補的なＰＮＡを演算用核酸にハイブリダイゼーションさせておいてもよい。この場合、ＤＮＡとＤＮＡのハイブリッドよりも、ＤＮＡとＰＮＡのハイブリッドは安定であるため、５’エクソヌクレアーゼ活性のあるポリメラーゼであっても除去することなできない。従って、Ｓ配列がストッパーとして機能する。
【０２７４】
また、マーカーオリゴヌクレオチドの蛍光標識の種類を増やしてもよい。現在は、数多くの蛍光色素が開発されており異なる核酸に異なる蛍光色素を標識することが可能である。そのようにすれば、同時に多くの演算用核酸を標識することができる。例えば、この実施の形態において、Ｍ_１マーカーオリゴヌクレオチドとＭ_２マーカーオリゴヌクレオチドとの間で蛍光分子を変えれば、検出した際に論理式の括弧で囲まれたどちらの条件が充足されたのかが判明する。また、演算用核酸毎にマーカー配列を変え、マーカーオリゴヌクレオチドに標識する蛍光分子も変えれば、１つのチューブに複数種類の演算用核酸を入れて同時に演算反応をすることもできる。更にまた、蛍光強度は演算核酸の表現する論理式の充足度に比例するので、その大きさを知ることも可能である。
【０２７５】
また、演算結果を得るに当たって、次のようなこともできる。即ち、この実施の形態における存在オリゴヌクレオチドと不存在オリゴヌクレオチドを増幅する工程において、ＰＣＲ反応を増幅が飽和するように十分なサイクル数行えば、「存在」を「１」、「不存在」を「０」とする２値の論理演算が可能である。一方、ＰＣＲ反応を増幅を飽和させず、元のｃＤＮＡの存在量に比例した量だけ得られるようにサイクル数を抑えれば、論理式を区間［０，１］で確率的に評価することができる。例えば、発現遺伝子の場合は発現量に応じた結果が得られ、ゲノム配列であれば、ヘテロ接合かホモ接合かの違いを演算結果で知ることができる。
【０２７６】
さらに演算用核酸をＤＮＡマイクロアレイのように、基板上に微小スポット状に固定し、その場所と演算用核酸の論理式が対応するようにアドレシングしておいてもよい。このようにしたときはマーカーオリゴヌクレオチドには蛍光標識をしておくのが好ましい。先に述べた演算反応をマイクロアレイ上の演算用核酸で行うと、ＤＮＡマイクロアレイの読みとり用スキャナで演算結果を読みとることができる。
【０２７７】
また、このマーカーオリゴヌクレオチドにビオチン分子を標識していてもよい。このとき、演算用核酸の５’端にはビオチンを付けず、３’端、５’端にクローニングのための制限酵素認識配列を含むものにする。さらに反応においては図３９に示すような演算用核酸をストレプトアビジン磁気ビーズにより捕獲する工程を行わない。この場合の検出反応はマーカーオリゴヌクレオチドをハイブリダイゼーション反応した後で、ストレプトアビジン磁気ビーズによって演算用核酸をハイブリダイズしたマーカーオリゴヌクレオチドもそうでないものも両方とも捕獲し、捕獲された演算用核酸をクローニングし、シーケンサーで塩基配列を読み取る。それにより演算結果が「１」となる演算用核酸を確認することができる。このようにすれば１つのチューブに複数種の演算用核酸を入れて反応を行うことができる。
【０２７８】
ここでは、４つの標的配列を用いた例を用いたが、更に多くの標的配列を対象とすることも可能である。また、ここでは、ビオチンとストレプトアビジンを回収を行うためのタグとして使用したが、これに限られるものではなく、タグとこれに対して高親和性を有するものであればどのような物質を使用してもよい。
【０２７９】
（ｂ）第２の実施の形態
本発明の好ましい態様に従うと、上述した方法において、ＤＣＮ配列として正規直交化配列を用いてもよい。「正規直交化配列」とは人工的に設計した塩基配列であって、「正規」とは融解温度（Ｔ_ｍ）が揃っていることを示し、「直交化」とはミスハイブリダイゼーションが起きず、自己分子内で安定な構造をとらないということを意味する。
【０２８０】
例えば、１５塩基の正規直交化配列を求めるには、任意の５塩基をランダムに生成する。これら短い塩基配列を「タプル」と呼ぶことにする。５塩基長のタプルは４^５＝１０２４種類ある。これらタプルの中から３つを選んで連結し１５塩基を構成する。この連結に用いたタプルに相補的なタプルは以降の連結には用いない。ここで、これらを連結した１５塩基の配列のＴ_ｍが±３℃以内に揃うような１５塩基のセットを作る。また、自己分子内で安定な構造を取るかどうかも計算し、安定な構造をなすならばそのような１５塩基は排除する。
【０２８１】
最後に全ての配列同士で互いに安定な２本鎖を形成しないかを検証する。以上の方法で生成した１５塩基の配列は反応温度を適切に選べば互いにハイブリッドを形成せず、混在しても独立したハイブリダイゼーション反応をするのでより好ましい。これら正規直交化配列を特定の遺伝子塩基配列と対応関係を持つように核酸ａ_ｉとＡ_ｉの配列を選び第１の実施の形態に従って反応を行えば、反応条件がより簡単になり、しかもより正確な演算反応を行うことが可能である。
【０２８２】
（ｃ）第３の実施の形態
続いて、本発明の好ましい態様に従うと、複数の遺伝子座にそれぞれ特定の塩基配列が存在することで決まるような遺伝子型を判定するためにも使用できる。ここでは、そのような遺伝子型を判定する方法の例を示す。
【０２８３】
遺伝子型に対応する論理式を設定し、その論理式に基づき演算用核酸を設計する。それにより、電子計算機や判定表を用いずに遺伝子型を判定することができる。具体的には、例えば遺伝子座１で塩基Ａ、遺伝子座２で塩基Ｔ、遺伝子座３で塩基Ｇであるとき遺伝子型Ａと判定でき、また、遺伝子座１で塩基Ａ、遺伝子座２で塩基Ｃ、遺伝子座３で塩基Ｔであるとき遺伝子型Ｂであり、また、遺伝子座１で塩基Ａ、遺伝子座２で塩基Ｃ、遺伝子座３で配列Ｇであるとき遺伝子型Ｃと判定できるならば、それぞれを満たす論理式は表４の右端欄に示す通りになる。
【０２８４】
【表４】

【０２８５】
遺伝子型判定はそれぞれの論理式と対応する演算用核酸により演算反応を行う。先ず、式の各要素について遺伝子座に特定の配列が存在する場合の値を１とし不存在の時を０とし、最終的に式全体の値が１となる式があるならば、その式に対応する遺伝子型が判定結果となる。このような核酸を用いた演算いる本発明の方法により、遺伝子座が多くあり且つ遺伝子型は極少ないような遺伝子の型判定が、複雑な表や電子計算機を必要としない簡便なものとなる。
【０２８６】
（ｄ）第４の実施の形態
更に、本発明の好ましい態様に従うと、上述の方法は、癌細胞の遺伝子発現において各種遺伝子がどのような発現、非発現の条件（即ち、状態）下にあるかのを調べることにも利用できる。上述した第１の実施の形態に対すると、第３の実施の形態は、所謂「逆問題」とも呼ぶことも可能である。
【０２８７】
予め、様々な論理式を表す演算用核酸を準備する。この演算用核酸は、５’端をリン酸化した各種の論理式要素、すなわちＤＣＮ_ｉ、ＤＣＮ_ｉ ^＊、Ｓ、Ｍ等からなる配列を有する。これらに加えて、例えば、図４５に示すような論理式要素を連結するための連結用相補核酸９を準備する。連結用相補核酸９の配列は、連結を所望する仕切部分に隣接して存在する２つの部分の配列に相補的な配列を有すればよい。
【０２８８】
これら核酸をハイブリダイゼーションし、ライゲースで連結反応すれば任意に論理式要素が結合された演算用核酸が得られる。連結用相補核酸の配列を十分考慮して設計することで論理式として不適当なものが生成されないようにできる。
【０２８９】
この演算用核酸を用いて逆問題を解くには、癌細胞から取得したｃＤＮＡを第１の実施の形態にしたがって、論理式要素の配列に変換し、あらかじめ準備しておいた様々な論理式を表す演算用核酸により論理演算を行う。最後にこれら演算用核酸が表現する論理式のうち、満たされるものがあるかどうかを検出する。このうち、論理式の値が１となった演算用核酸の意味する内容を解釈することによって、どのような遺伝子の発現、非発現状態が満たされる条件にあるのかを解明することが可能である。或いは少なくともその部分条件を解明することが可能である。
【０２９０】
任意に作製した演算用核酸の配列を同定するには、先ず、１つの容器において反応する。次に、第１の実施の形態のビオチン分子を結合したマーカーオリゴヌクレオチドにてストレプトアビジン磁気ビーズに演算用核酸を回収する。続いて、シーケンシングを行い、演算分子の内容を読みとる方法で論理式を読み取ることが好ましい。また、任意に演算用核酸を作製せずとも、論理式を決めて連結により演算用核酸を合成する方法でおこなってもよい。この場合には、それら論理式をＤＮＡマイクロアレイにアドレシングして固定し、検出時に論理式を確認してもよい。または、１つの容器に１種類の演算用核酸を入れて反応を行っても良い。
【０２９１】
これらの結果を正常細胞と比較すれば、未知のゲノム塩基配列の組み合わせにより生ずる遺伝病や、未知の遺伝子発現の組み合わせによって生ずる遺伝子異常による癌などの疾病の原因遺伝子を容易に特定することができる。
【０２９２】
（ｅ）考察
従来の技術には、特定の核酸が存在しないことを核酸表現に変換できる技術はなく、また、そのような思想すらない。例えば、ｍＲＮＡの発現状態を検出するにはＤＮＡマイクロアレイがよく用いられている。このマイクロアレイには既知の遺伝子塩基配列をもとに設計したオリゴＤＮＡ、またはあらかじめ取得したｃＤＮＡをプローブとしてスライドガラス上にアレイ状に固定している。
【０２９３】
このマイクロアレイで発現を検出するには、ｍＲＮＡから蛍光標識したｃＤＮＡを作製し、マイクロアレイのプローブと該ｃＤＮＡとをハイブリダイゼーション反応をさせ、特定の配列のプローブを固定した場所に特定の標識したｃＤＮＡが結合して光ることで検出する。ところが、発現していないｍＲＮＡからは標識ｃＤＮＡが作製できないので、遺伝子が発現していないことを検出できない。すなわち、実験中にｍＲＮＡが失われたり、遺伝子が発現しているにもかかわらず生成される蛍光標識ｃＤＮＡが少ないために不存在であると判定されることがある。このような従来の方法とは異なり、本発明の１態様に従うと、不在の核酸の情報を可視化することが可能である。
【０２９４】
また、遺伝子核酸を反応させるとき、数多くの遺伝子が混在した状態では思わぬ核酸同士がハイブリッドを形成して反応することがある。また、塩基配列に含まれるグアニンやシトシンなどの塩基の数は２本鎖核酸の構造を安定化する。このような塩基が、取り扱う遺伝子核酸毎にまちまちであれば、ハイブリッドを形成する最適な温度が異なる。そのため、全ての核酸がミスマッチのない、適切なハイブリッドを形成できるとは限らない。また、自己分子内で構造をとり、標的配列との反応性が低くなるような塩基配列をもつ核酸が混在している場合には、理論的に予想される反応が進まないこともあり得る。このような従来の方法に比較して、本発明の態様に従う方法では、情報を、好ましい条件で設計した核酸分子に、置き換えてから、反応に使用するので、安定した反応を行うことが可能である。
【０２９５】
また、化学発光など酵素を用いた検出法は高感度であるが、検出のための処理が面倒で時間がかかる。更に、１チューブ内で複数の種類の化学発光を行うことは難しい。また、演算用核酸を用いて演算した場合、その結果を見るのに単一の発色もしくは発光反応では１チューブで１種類の演算用核酸しか反応できない。これに対して、本発明の態様に従えば、感度よく、多数の標的核酸を同時に感度よく検出することが可能である。
【０２９６】
また、従来の方法では、核酸の存在条件のみに基づいて作った論理式が演算用核酸に書き換えられている。ところがおよそ世の中に存在する問題は所謂「逆問題」である。例えば遺伝病における各遺伝子の発現状態について調べたとき、どのような遺伝子核酸の存在、不存在の条件が満たされれば病気になるのかが問題である。従って、遺伝子核酸の存在および不存在の論理式を求めることこそが問題なのである。従来では、このような問題は、ＤＮＡマイクロアレイのデータを大型計算機によりクラスタ解析して解いていた。従って、非常に多くの時間と費用とが必要とされていた。本発明の態様に従うと、短時間に、且つ経済的にそのような問題を解くことが可能である。
【０２９７】
ここでは、遺伝子解析のための方法についても示したが、これに限定されず、本発明の範囲を超えることなく種々の並列計算による情報処理を行うことが可能である。即ち、遺伝子解析以外の情報の、例えば、数学的な問題を並列処理を行って解く場合においても、優れた利点を得られる当業者には容易に理解されるであろう。
【０２９８】
ＩＩＩ．その他の適用例
上述したように本発明の１態様は、
分子の化学反応により分子計算を実行する分子計算装置であって、
分子の化学反応の反応制御を行う反応制御部と、前記反応制御部の反応制御に基づき分子の化学反応を生じさせ、該化学反応の反応結果を検出する自動検出部とを備えた分子計算部と、
分子を核酸配列に基づき符号化した符号分子に対応付けて定義された変数及び定数と、前記符号分子の演算反応に対応付けて定義された関数により記述されたプログラムを、前記符号分子の演算反応に対応する分子の化学反応の反応制御を行う前記反応制御部を駆動させるための制御命令に変換して前記制御命令の手順を生成する制御命令生成部と、
前記制御命令の手順を出力する出力部とを有する電子計算部
を備える分子計算装置
を提供する。
【０２９９】
また、この態様では、分子の化学反応のうち、核酸配列の反応の規則性に着目し、その規則性を演算反応に対応付けた。核酸配列の反応の規則性のみならず、分子の他の生物学的及び生化学的な規則性に基づき生じる化学反応を演算反応に対応付けることにより、同様の分子計算が可能となる。
【０３００】
例えば、本発明の核酸配列以外の生物学的および生化学的物質、例えば、ペプチド、オリゴペプチドおよびポリペプチド、蛋白質、並びに抗原および抗体などの生物学的および生化学的な要素を用いることができる。この場合、これら要素を核酸配列に替えて符号化し、符号化されたこれら各要素に対して対応付けて定義された変数および定数と、符号化された要素の演算反応に対応付けて定義された関数により記述されたプログラムにより原始プログラムを作成すればよい。
【０３０１】
この明細書には、以下の発明が含まれることを確認する。
【０３０２】
分子の化学反応により分子計算を実行する分子計算の計画を設定する分子計算計画設計装置であって、
分子を該分子の生物学的性質又は生化学的性質に基づき符号化した符号分子に対応付けて定義された変数及び定数と、前記符号分子の演算反応に対応付けて定義された関数により記述されたプログラムを、前記符号分子の演算反応に対応する分子の化学反応の反応制御を行う反応制御部を駆動させるための制御命令に変換して前記制御命令の手順を生成する制御命令生成部と、
前記制御命令の手順を出力する出力部。
【０３０３】
ＩＶ．分子計算部の適用例
図３に示した分子計算部２２の構成は一例にすぎない。
【０３０４】
例えば、以下の（１）〜（１０）に示される全自動処理に適した検出装置を分子計算部２２に適用することにより、省力化と同時に処理速度を向上させることができる。上述した各実施形態と同一の構成には同一符号を付し、詳細な説明は省略する。
【０３０５】
（１）ＦＣＳを用いた検出装置
ＦＣＳ（Ｆｌｕｏｒｅｓｃｅｎｃｅ Ｃｏｒｒｅｌａｔｉｏｎ Ｓｐｅｃｔｒｏｓｃｏｐｙ：蛍光相関分析法）を用いた検出装置は、例えば特開２００１−２６９１９８や特開２００２−２８１９６５に示される。
【０３０６】
図４６はＦＣＳを用いた共焦点顕微鏡による検出装置の一例を示す図である。レーザ光源４１から励起光を照射し、ミラー５４ａ及び５４ｂで反射させ、フィルタ４２（ＩＦ）を通過させた後、このレーザ光を集光レンズ４３で集光し、ダイクロイックミラー４４（ＤＭ）によって試料４５中の一点に照射する。これにより、試料４５中の蛍光物質がレーザ光で励起され、蛍光発光する。この蛍光発光は、フィルタ４６（Ｆ）、レンズ４７、ミラー５４ｃ、レンズ４８を介してピンホール４９を通過する。ピンホール４９を通過したことにより、試料４５中の焦点中の蛍光物質から発せられた蛍光のみが光増倍管５０（ＰＭＴ）で増幅される。増幅信号は増幅器５１、弁別回路５２を介してデジタル相関器５３に出力される。デジタル相関器５３は、受信した蛍光信号をＦＣＳにより解析する。
【０３０７】
この検出装置により、平均数個（場合により１個）の蛍光物質が発する蛍光の強度を一定時間測定し、ブラウン運動に由来する蛍光のゆらぎの自己相関関数をとり、この自己相関関数を分析することにより、蛍光物質に関する種々のデータを取得する。
【０３０８】
例えば、レーザ光源４１、光増倍管５０、弁別器５２、デジタル相関器５３などを自動制御部１５１で制御し、また試料４５を自動制御部１５１で制御される搬送機構１６０を用いて搬送する。デジタル相関器５３で得られた検出結果は、検出部１５８の検出結果として通信部２４を介して、電子計算部２１に送られる。
【０３０９】
例えば試料４５が、底面がガラス製の複数のウェル４５ａを有するマルチウェルプレートの各ウェル４５ａに滴下される構成により、反応容器と検出容器を同一のウェルを用いて実現される。これにより、高感度の検出装置の自動化が容易にできる。
【０３１０】
図４７はＦＣＳの解析原理を示す図である。
【０３１１】
図４７（ａ）に示すように、微小な測定対象領域５５にレーザ光を照射する。この微小領域中に存在する蛍光物質から、発せられる蛍光の蛍光強度を経時的に測定し、例えば図４７（ｂ）や（ｃ）に示されているようなデータを取得する。そして、異なる２時点の蛍光強度Ｉ（ｔ）とＩ（ｔ＋τ）の積の期待値を計算し、自己相関関数Ｇ（τ）＝＜Ｉ（ｔ）Ｉ（ｔ＋τ）＞を得る。
【０３１２】
そして、得られた自己相関関数Ｇ（τ）を解析し、各核酸プローブについて、添加前と添加後の自己相関関数を比較する。
【０３１３】
このような分析の概念は、図４７（ｂ）及び（ｃ）により明確になる。
【０３１４】
図４７（ｂ）に示すように、蛍光強度の関数Ｉ（ｔ）は、核酸プローブが標的対立遺伝子に結合していない場合には、分子のサイズが小さいので、ブラウン運動の速度が大きく、Ｉ（ｔ）の周波数が大きい。
【０３１５】
これに対して、核酸プローブが標的対立遺伝子に結合すると、図４７（ｃ）に示すように、周波数の小さいデータが得られる。従って、蛍光強度を基にして得られた自己相関関数を比較することにより、プローブが結合したか否かが明らかになる。このＦＣＳによっていずれの核酸プローブが結合したのかを決定するためには、例えば各核酸プローブを励起波長や蛍光波長の異なる蛍光標識で区別すればよい。
【０３１６】
また、各核酸プローブのサイズが異なれば、ブラウン運動の変化及び自己相関関数も異なる。従って、サイズの異なる核酸プローブを用いることによっていずれの核酸プローブが結合したかを決定できる。
【０３１７】
（２）ＤＮＡキャピラリアレイを用いた検出装置
ＤＮＡキャピラリアレイを用いた検出装置は、例えば特開平１１−７５８１２に示される。
【０３１８】
図４８はＤＮＡキャピラリ７０の一例を示す図である。図４８に示すように、例えばガラス製の光透過性でかつ円筒状の注入用開口部７４及び排出用開口部７５を両端に有するキャピラリ７１の内壁７２に、流路に沿って縞状に種々のＤＮＡプローブ７３ａ，７３ｂ，７３ｃ…がキャッピング剤の光反応によって順次固定化されている。測定を行うには、蛍光標識された一本鎖ＤＮＡを含む試料をキャピラリ７１の注入用開口部７４から導入して、各ＤＮＡプローブ７３ａ，７３ｂ，７３ｃ…と反応させた後、蛍光を測定する。キャピラリ７１の光る位置を計測することで、試料中に含まれるＤＮＡの配列を決定できる。
【０３１９】
図４９はＤＮＡキャピラリ７０を集積化したキャピラリアレイ７５を示す。図４９（ａ）は斜視図、（ｂ）は上面図である。
【０３２０】
ガラス又はシリコン等の材料からなる基板８１には、図示するようなパターンの溝が設けられており、これにより複数のＤＮＡキャピラリ８２ａ，８２ｂ，８２ｃ…が形成されている。各々のＤＮＡキャピラリ８２ａ，８２ｂ，８２ｃ…の一端は個別出入口８３ａ，８３ｂ，８３ｃ…の真下まで延在して、各個別出入口８３ａ，８３ｂ，８３ｃ…を通じて外気に開放されており、他端は合流路としての連結流路８６で一本に連結され、共通出入口８４の真下まで延在し、この共通出入口８４を通じて外気に開放されている。個別出入口８３ａ，８３ｂ，８３ｃ…及び共通出入口８４は、例えばスクリーン印刷技術により接着剤を薄膜状に接着することにより取付可能である。
【０３２１】
ＤＮＡプローブ８５ａ，８５ｂ，８５ｃ…は、図示するように各々のＤＮＡキャピラリ８３ａ，８３ｂ，８３ｃ…の流路に沿って流体の流れる方向とは直角な同一直線８７_１，８７_２，８７_３，…上に配置されている。各個別出入口８３ａ，８３ｂ，８３ｃ…から異なる試料を同時に注入し、共通出入口８４へ排出する。例えば９６穴試料保持容器がアレイ状に設けられた試料保持アレイから、各保持容器に対応して設けられたホース状の注入機構により各個別出入口８３ａ，８３ｂ，８３ｃ…への試料の注入を行う。これにより、多種類の試料を同時に処理することができる。また各キャピラリに反応液を注入するだけでハイブリダイゼーション反応ができるので、並列処理による高速化、自動化ができる。例えば、ホース状の注入機構の各個別出入口８３ａ，８３ｂ，８３ｃ…への配置を自動制御部１５１の制御の下で搬送機構１６０により行う。さらに、各々のＤＮＡキャピラリ８３ａ，８３ｂ，８３ｃ…に対応させて蛍光検出装置を検出部１５８として設け、この蛍光検出装置が各ＤＮＡキャピラリ８３ａ，８３ｂ，８３ｃ…について同時に流路に沿ってスキャンしていく。スキャン位置と、蛍光の検出位置を対応付けて解析することにより、どのキャピラリのどのプローブが発光しているかを自動で検出できる。解析結果は、検出部１５８から通信部２４を介して電子計算部２１に送信される。これにより、検出までを含め他種類の試料を同時に処理することが出来る。
【０３２２】
（３）プレート内マイクロアレイを用いた検出装置
プレート内マイクロアレイを用いた検出装置は、例えば特表２００１−５１０３３９に示される。
【０３２３】
図５０はプレート内マイクロアレイを用いた検出装置９０の構成の一例を示す図である。多数の穴がアレイ状に設けられ反応容器９１を構成し、この反応容器９１が複数設けられた多穴プレート９２の各々の底部に、予め配列が既知の異なるＤＮＡプローブ９５を固相化する。そして、各反応容器９１内に、蛍光標識された一本鎖ＤＮＡ分子を含む溶液を注入し、ハイブリダイズさせる。そして、どの位置が光っているかを蛍光検出機構９３により検出する。これにより、溶液中に含まれるＤＮＡの配列を決定することができる。
【０３２４】
この蛍光検出機構９３は、図３の検出部１５８に対応する。蛍光検出機構９３は、基体９３ａと蓋部９３ｂからなる。基体９３ａには、多穴プレート９２を配置するプレート配置部が設けられている。また、プレート配置部の下面には、ＣＣＤ９３ｃが配置されている。このＣＣＤ９３ｃ上に多穴プレート９２を配置し、蓋部９３ｂを閉じて蓋部９３ｂに設けられた照明装置９３ｄから光を照射すると、ハイブリダイズした反応容器９１が発光する。この発光を、ＣＣＤ９３ｃにより検出し、基体９３ａ内のＣＣＤイメージャ９３ｅで解析する。解析結果は図示しない通信部２４を介して電子計算部２１に送信される。ＣＣＤ９３ｃ、照明装置９３ｄは、各反応容器９１毎にアレイ状に２次元配置されてもよいし、例えば図５０に示すように、アレイ中の一列分を検出可能に１次元的に配置され、駆動機構によりレール９６上をこれらＣＣＤ９３ｃ、照明装置９３ｄをスキャンさせてすべての反応容器９１を検出するように構成されてもよい。例えばＣＣＤ９３ｃ、照明装置９３ｄ、ＣＣＤイメージャ９３ｅは、自動制御部１５１により自動制御される。
【０３２５】
上記実施形態における反応容器と同じマルチウェルプレート、例えば９６穴のものを使用した場合、反応後すぐに検出出来るので、処理を高速化できる。
【０３２６】
（４）ＥＣＡ（Ｅｌｅｃｔｒｏｎｉｃ Ｃｈｅｍｉｃａｌ Ａｒｒａｙ）を用いた検出装置
ＥＣＡ（Ｅｌｅｃｔｒｏｎｉｃ Ｃｈｅｍｉｃａｌ Ａｒｒａｙ）を用いた検出装置は、例えば特開平６−２４５８００、特開平５−２８５０００、特開平５−１９９８９８、特開２０００−８３６４７、特開２００２−３００８７１、特開２００２−２１４１９２などに示される。
【０３２７】
図５１はＥＣＡチップ１００の構成の一例を示す図である。ＥＣＡチップ１００は例えば８０塩基の合成オリゴＤＮＡプローブ１０１を金等からなる微小電極１０２上に共有結合させたセラミックなどにより形成されたチップである。
【０３２８】
図５２はＥＣＡチップを用いた検出原理を説明するための図である。図５２に示すように、合成オリゴＤＮＡプローブ１０１が固定化された固定プローブ電極１０２において、サンプル溶液と接触させることによりサンプルＤＮＡ１０３を合成オリゴＤＮＡプローブ１０１と結合させ二重鎖ＤＮＡを生成する。この二重鎖に、インターカレータ１０４を取り込ませると、対合したオリゴＤＮＡと結合した電極１０２にだけ電流が流れ（１０３）、結合しない電極１０２には電流が流れず（１０４）、これを例えば各電極１０２に接続された検出部１５８により電気的に検出し、通信部２４を介して電子計算部２１に送信する。検出部１５８による検出動作は、例えば自動制御部１５１により制御される。
【０３２９】
このように、電気信号として対合反応を検出できるため、レーザスキャン装置が不要である。また、オリゴＤＮＡが微小電極１０２上に貼り付いており、また二本鎖形成量とインターカレータ１０４の量には相関関係が得られるため、これにより、ＤＮＡの定量的検出が可能となる。
【０３３０】
このように、ＥＣＡチップを用いた検出装置により、無標識で高速に検出することができる。また、高感度でＰＣＲが不要になり、そして再使用が可能な低コスト製品であり、チップコストが低減できる。
【０３３１】
（５）電極固定ＤＮＡプローブアレイを用いた検出装置
電極固定ＤＮＡプローブアレイを用いた検出装置は、例えば特表平９−５０４９１０、特表平９−５０３３７７に示される。
【０３３２】
図５３は電極固定ＤＮＡプローブアレイを用いた検出装置の検出手法を示す図である。まず、図５３（ａ）に示すように、基板１０６に形成された複数の電極１０７にＤＮＡプローブ１０８ａ，１０８ｂ，１０８ｃを固定化する。この段階では、電極１０７には電位は与えられていないか、あるいはマイナスの電位が与えられている。そして、図５３（ｂ）に示すように、各電極１０７を自動制御部１５１により制御される電圧源１０９によりプラス電位にし、マイナス（−）電荷を持つＤＮＡ分子を引きつけ、ハイブリダイゼーション効果を高める。ハイブリダイゼーション反応の後、図５３（ｃ）に示すように各電極１０７を電圧源１０９によりマイナス（−）に保持し、ＤＮＡプローブ１０８ａ，１０８ｂ，１０８ｃに強く結合しているＤＮＡ以外のものを排除し、検出を行う。検出は、予め溶液中のＤＮＡ分子に蛍光標識などをほどこしておき、それを蛍光検出機構などの検出部１５８により計測する。
【０３３３】
このように、電極固定ＤＮＡプローブアレイを用いた検出装置によれば、単なるマイクロアレイと違って、高速にハイブリダイズ反応出来る。
【０３３４】
（６）マイクロビーズアレイを用いた検出装置
光学的に識別可能なマイクロビーズアレイを用いた検出装置は、例えば特表２００１−５２０３２３、特表２００２−５３４６５７などに示される。
【０３３５】
図５４はマイクロビーズアレイを用いた検出装置の検出原理を示す図である。
【０３３６】
まず、図５４（ａ）に示すように、光学的に個別識別可能な第１の色のマイクロビーズ１１１に配列が既知のプローブＤＮＡ１１２を固定化させた溶液が含まれるマイクロチューブ１１４を用意する。この際、マイクロビーズ１１１の第１の色と、固相化されたプローブＤＮＡ１１２を対応させておく。そして、このマイクロチューブ１１４内の溶液に第２の色で蛍光ラベルされたサンプルＤＮＡ分子１１３を混合させ、サンプルＤＮＡ分子１１３とプローブＤＮＡ１１２をハイブリダイズさせる。次に、図５４（ｂ）に示すように、この溶液をマイクロチューブ１１４の先端から１滴ずつ滴下し、落下する溶液に第１レーザ照射源１１５から第１レーザ光を照射し、これを光センサ１１７を用いて検出してマイクロビーズ１１１の直径及び色を測定するとともに、第２レーザ照射源１１６から滴下溶液に第２レーザ光を照射させて第１レーザ光に交差させ、光電子増倍管１１８により検出することによりマイクロビーズ１１１表面の蛍光量を測定する。測定信号は検出部１５８、通信部２４を介して電子計算部２１に送信される。第１レーザ光によりマイクロビーズ１１１の有無を検出してどのようなプローブＤＮＡ１１２なのかを検出し、第２レーザ光でサンプルＤＮＡ分子１１３のハイブリダイズの有無を検出することができる。第２レーザ光でサンプルＤＮＡ分子１１３が観測されなければ、ハイブリダイズされていないことを示すため、これにより、溶液中のＤＮＡ分子の配列を決定することができる。第１レーザ照射源１１５、光センサ１１７、第２レーザ照射源１１６及び光電子増倍管１１８、検出部１５８は、自動制御部１５１により制御される。
【０３３７】
また、このマイクロビーズアレイを用いた検出装置によれば、フロー系で検出出来るので、高速高感度に検出出来る。
【０３３８】
（７）マイクロトランスポンダを用いた検出装置
個別に識別可能なマイクロトランスポンダを用いた検出装置は、例えばＵＳ ６０４６００３に示される。
【０３３９】
図５５はマイクロトランスポンダを用いた検出装置１２０の一例を示す図である。図５５に示すように、金属コイルアンテナ１２１はフローサイトメータのフローセル１２２の周囲に覆われている。個別識別可能なトランスポンダ１２３はフローセル１２２を通過し、スキャナ装置１２４でデコードされる。トランスポンダ１２３から送られるデータを搬送する信号は増幅器で増幅され、スキャナ装置１２４で処理される。トランスポンダ１２３がデコードされるとき、レーザ照射源１２０ａからのレーザ光により、トランスポンダ１２３からの蛍光がフローサイトメータで検出及び解析される。これらスキャナ装置１２４及びフローサイトメータの処理結果及び解析結果は、通信部２４を介して電子計算部２１に送信される。レーザ照射源１２０ａ、スキャナ装置１２４は、自動制御部１５１により制御される。
【０３４０】
このように、マイクロトランスポンダを用いた検出装置１２０によれば、フロー系で検出出来るので、高速高感度に検出出来る。
【０３４１】
（８）周波数変換素子を用いた検出装置
周波数変換素子を用いた検出装置は、例えば特開平６−９４５９１に示される。
【０３４２】
図５６は周波数変換素子を用いた検出装置１２５の一例を示す図である。周波数変換素子１２７に設けられた電極１２６上に、プローブＤＮＡ１２６ａを固相化しておく。そして、この電極１２６をサンプル溶液中に浸漬し、サンプル溶液中のサンプルＤＮＡ１２６ｂとハイブリッドを形成させる。そのハイブリッド形成による重量変化に伴う周波数変換素子１２７の周波数変化を検出器１２８により検出し、ハイブリッドの有無を検出する。検出器１２８の検出結果は、通信部２４を介して電子計算部２１に送信される。検出器１２８は自動制御部１５１により制御される。
【０３４３】
このように、周波数変換素子を用いた検出装置１２５によれば、無標識で高感度に検出出来る。
【０３４４】
（９）表面プラズモン共鳴を用いた検出装置
表面プラズモン共鳴（Ｓｕｒｆａｃｅ Ｐｌａｓｍｏｎ Ｒｅｓｏｎａｎｃｅ）を用いた検出装置は、例えば特開２００２−５１７７４に示される。
【０３４５】
図５７は表面プラズモン共鳴を用いた検出装置１３０の一例を示す図である。センサチップ１３１の裏面に、プローブＤＮＡ１３２を固定化しておく。そして、このセンサチップ１３１裏面に、フローセル１３５により検体ＤＮＡ１３３を導入し、プローブＤＮＡ１３２と検体ＤＮＡ１３３をハイブリダイズさせる。この状態で、プローブＤＮＡ１３２が固定化されていない側のセンサチップ１３１表面に形成された金属膜１３１ａに光源１３４から偏光をプリズム１３４ａを介して全反射するように照射する。そして、センサチップ１３１表面で反射した反射光を検出部１５８で検出する。検出部１５８の検出結果は通信部２４を介して電子計算部２１に送信される。光源１３４、検出部１５８は自動制御部１５１により制御される。
【０３４６】
検出された反射光の一部には、反射光強度が低下した部分が観察される。この光の暗い部分の現れる角度は、センサチップ１３１表面での質量に依存する。リガンドに検体ＤＮＡ１３３を添加し結合反応が起きると、質量変化が生じ、光の暗い部分が図のＩからＩＩにシフトする。解離により質量が減少すれば、逆にＩＩからＩにシフトする。この検出装置１３０では、図のＩからＩＩにシフトする量、すなわちセンサチップ１３１表面での質量変化を縦軸にとり、質量の時間変化を測定データとして取得する。これにより、プローブＤＮＡを標識することなく、プローブＤＮＡと検体ＤＮＡの相互作用をリアルタイムで高感度に検出できる。
【０３４７】
なお、同一のチップ１３１に対して複数の異なる複数のプローブＤＮＡ１３２を固相化して、同一のサンプル溶液と反応させた後に、測定したいプローブのみを選択的に順次光照射して測定すれば、複数の測定データを同時に解析することも可能となる。
【０３４８】
（１０）糸チップを用いた検出装置
図５８は糸チップを用いた検出装置１４０の構成の一例を示す図である。糸状部材１４１の複数の位置に、プローブＤＮＡ１４２を固相化する。そして、このプローブＤＮＡ１４２の固相化された糸状部材１４１を芯１４３に巻き付ける。この糸状部材１４１が巻き付けられた芯１４３をサンプル溶液１４５を含むマイクロチューブ１４４に入れ、ハイブリダイズさせ、洗浄させる。この洗浄したマイクロチューブ１４４に光源１４６から光を照射し、その反射光を検出部１５８で検出する。検出結果は通信部２４を介して電子計算部２１に送信される。なお、検出結果をいったんコンピュータにより解析し、解析結果を電子計算部２１に送信してもよい。光源１４６及び検出部１５８は、自動制御部１５１により制御される。
【０３４９】
このように、プローブＤＮＡを糸状部材１４１に巻き付けることで、プローブＤＮＡの密度を高めることができる。また、検出装置としての全自動化も容易である利点がある。
【０３５０】
ＤＮＡプローブを固相化する代わりに、複数の異なる患者由来の検体ＤＮＡを固相化するようにしてもよい。特に、糸巻きの軌跡に沿って測定データが整列するので、取り違いも生じにくく、アドレス毎の高速解析が可能となる。
【０３５１】
本実施形態は、（１）〜（１０）に示される検出装置に限定されない。（１）〜（１０）に示される検出装置を組み合わせて用いることもできる。
【０３５２】
例えば、（２）に示すＤＮＡキャピラリと（６）に示すマイクロビーズを組み合わせて用いることもできる。図５９はマイクロビーズをキャピラリに配置したビーズキャピラリアレイの一例を示す。（ａ）は上面図、（ｂ）は断面図である。具体的には、キャピラリ１４７中に、複数のマイクロビーズ１１１が整列配置されている。このマイクロビーズ１１１が成立されたキャピラリ１４７にサンプル溶液を導入し、光学顕微鏡で観察することにより、サンプル中のサンプルＤＮＡの有無を検出することができる。このように、（２）に示すキャピラリと、（６）に示すマイクロビーズを組み合わせた検出装置が提供される。また、（１）〜（１０）の他の組合せの適用も可能であり、例えば（１）と（２）、（２）と（１０）、（２）と（６）、（３）と（９）、（４）と（６）、（４）と（１０）、（７）と（１０）、（２）と（４）と（６）、（２）と（７）と（１０）という好適な組合せが挙げられる。
【０３５３】
【発明の効果】
以上詳述したように本発明によれば、分子計算機の高い並列計算性を活かし、しかも分子計算機だけでは実現することが難しい機能を電子計算機により補完することにより、従来の電子コンピュータよりも高速に計算を実行することが可能となる。
【図面の簡単な説明】
【図１】本発明の分子計算装置を示すブロック図。
【図２】本発明の分子計算装置の各部の配置を示す模式図。
【図３】本発明の分子計算装置を示すブロック図。
【図４】本発明の分子計算方法の処理のフローチャート。
【図５】分子計算方法の処理の変形例を示すフローチャート。
【図６】分子計算方法の処理の変形例を示すフローチャート。
【図７】分子計算方法の処理の変形例を示すフローチャート。
【図８】分子計算方法の処理の変形例を示すフローチャート。
【図９】分子計算方法の処理の変形例を示すフローチャート。
【図１０】分子計算方法のデータフロー図。
【図１１】手続または関数−演算反応変換データテーブルの一例を示す図。
【図１２】演算反応手順テーブルの一例を示す図。
【図１３】実行手順テーブルの一例を示す図。
【図１４】制御命令手順テーブルの一例を示す図。
【図１５】コマンドの処理の概要を示すフローチャート。
【図１６】コマンドの処理の概要を示すフローチャート。
【図１７】コマンドの処理の概要を示すフローチャート。
【図１８】コマンドの処理の概要を示すフローチャート。
【図１９】配列の同定方法の例を示す概念図。
【図２０】プログラムの流れを示すフローチャート。
【図２１】配列同定の結果を示すチャート。
【図２２】遺伝子解析のためのエンコード反応とデコード反応とを示すフローチャート。
【図２３】遺伝子解析のための分子設計例を示す図。
【図２４】遺伝子検出反応工程における各分子の状態を示す模式図。
【図２５】標的が存在した場合の反応系における各分子の状態を示す模式図。
【図２６】ストレプトアビジン磁気ビーズによる捕獲工程における各分子の状態を示す模式図。
【図２７】ＤＣＮ_ｉの抽出工程における各分子の状態を示す模式図。
【図２８】抽出工程で得られたＤＣＮ_ｉに相補的な配列の増幅の模式図。
【図２９】図２８の増幅により得られた増幅産物を捕獲する工程における各分子の状態を示す模式図。
【図３０】熱変性により一本鎖化する工程における各分子の状態を示す模式図。
【図３１】発現遺伝子の情報を存在分子に変換する工程における各分子の挙動を示す模式図。
【図３２】非発現遺伝子を検出し、不存在分子に変換するための最初の工程における反転オリゴヌクレオチドと存在分子の反応を示す模式図。
【図３３】ある非発現遺伝子のための反転オリゴヌクレオチドの状態を示す模式図。
【図３４】反転オリゴヌクレオチドの抽出工程における分子の状態を示す模式図。
【図３５】ストレプトアビジンを固定した磁気ビーズによるＤＣＮ_ｋ＊の捕捉とハイブリダイゼーションによる抽出工程における各分子の状態を示す模式図。
【図３６】演算用核酸を示す模式図。
【図３７】演算用核酸と存在分子および不存在分子とのハイブリダイゼーション工程における各分子の状態を示す模式図。
【図３８】図３７の工程の後に演算用核酸にハイブリッドした前記分子を伸長する工程における各分子の状態を示す模式図。
【図３９】マーカーオリゴヌクレオチドＭ_１およびＭ_２による計算結果の検出工程における各分子の状態を示す模式図。
【図４０】不存在分子の増幅のために不存在分子を抽出回収する工程における各分子の状態を示す模式図。
【図４１】不存在分子の増幅のためのＰＣＲ工程における各分子の状態を示す模式図。
【図４２】図４１の工程により生じた増幅産物を捕獲する工程における各分子の状態を示す模式図。
【図４３】図４２の工程で回収された増幅産物を１本鎖にする工程における各分子の状態を示す模式図。
【図４４】図４３の工程の１本鎖に対する不存在分子のハイブリダイゼーションを行う工程における各分子の状態を示す模式図。
【図４５】演算用核酸のランダムライブラリの作成方法において使用する連結用相補核酸と、連結対象となる演算用核酸の一部分を示す模式図。
【図４６】ＦＣＳを用いた検出装置の一例を示す図。
【図４７】ＦＣＳの解析原理を示す図。
【図４８】ＤＮＡキャピラリの一例を示す図。
【図４９】キャピラリアレイの一例を示す図。
【図５０】プレート内マイクロアレイを用いた検出装置の構成の一例を示す図。
【図５１】ＥＣＡチップの構成の一例を示す図。
【図５２】ＥＣＡチップを用いた検出原理を説明するための図。
【図５３】電極固定ＤＮＡプローブアレイを用いた検出装置の検出手法を示す図。
【図５４】マイクロビーズアレイを用いた検出装置の検出原理を示す図。
【図５５】マイクロトランスポンダを用いた検出装置の一例を示す図。
【図５６】周波数変換素子を用いた検出装置の一例を示す図。
【図５７】表面プラズモン共鳴を用いた検出装置の一例を示す図。
【図５８】糸チップを用いた検出装置の構成の一例を示す図。
【図５９】マイクロビーズをキャピラリに配置したビーズキャピラリアレイの一例を示す図。
【符号の説明】
１１…入力部、１２…入出力制御部、１３…記憶部、１４…演算部、１４ａ…翻訳・計算計画立案・実行部、１４ｂ…核酸配列計算部、１４ｃ…結果解析部、１５…演算部、１６…記憶部、１７…入出力制御部、１８…入力部、１９…出力部、２０…出力部、２１…電子計算部、２２…分子計算部、２３，２４…通信部、３０…核酸合成装置、１５１…自動制御部、１５２…ＸＹＺ制御ピペッタ、１５３…サーマルサイクラ反応容器、１５４…ビーズ容器、１５５…酵素容器、１５６…緩衝液容器、１５７…核酸容器、１５８…検出部、１５９…温度制御手段、１６０…搬送機構[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a novel molecular computer using a nucleic acid molecule.
[0002]
[Prior art]
Computers using semiconductor silicon have greatly improved in performance since their birth, and have performed complex calculations at low cost, contributing to humankind. These computers using semiconductor silicon are Neumann-type computers that mainly perform calculations using binary values of 0 and 1.
[0003]
A well-known problem to be studied in the field of computer science is the NP complete problem. Examples of such NP-complete problems include the traveling office worker problem and the prediction of the three-dimensional structure of proteins. There are two typical conventional methods for completely solving such a problem. One is to apply all possible solutions to the problem and verify whether the problem can be solved. Another is a method of solving a problem by finding an approximate solution.
[0004]
In the former method, the computation time required to obtain an answer increases exponentially in proportion to the size of the problem. The latter method has been proposed to perform such calculations at high speed. Several algorithms for finding approximate solutions have been proposed. These algorithms do not always find an exact solution and may overlook the solution.
[0005]
In order to obtain all the solutions at high speed by the former method, it is conceivable in the current technology to calculate a large number of parallelized computers in parallel. However, as the number of computers increases, power consumption increases, and a wider installation place is required. In addition, there are many technical issues such as how to make computers communicate with each other in order to parallelize the computers, and what connection method to use.
[0006]
On the other hand, a new computer paradigm called DNA computing was proposed by Adleman in 1994 to solve a problem that is difficult to solve with a Neumann-type computer (Science, 266, 1021- 4).
[0007]
Eidelman's idea is that a DNA molecule is regarded as a tape on which information is recorded, and an enzyme is used as a main body of information processing to execute the computation process of the Turing machine using the DNA molecule. That is, in order to solve a small-scale Hamiltonian pathway problem, Edelman uses DNA to form DNA corresponding to the pathway. Then, a method of selecting a solution DNA from the DNA was used. In addition, there is a report by Guanieri et al. (Science, 273, 220-3) that addition is performed using DNA molecules. In this way, the search for the possibility of operation using molecules has been performed.
[0008]
It has been found that the use of DNA for computation is advantageous in the following respects. For example, 1 pmol of a short DNA molecule of about several tens of bases (= 10^-12mol) readily dissolves in 100 μL of buffer solution. In addition, the number of molecules in the lysis solution is about 6 × 10¹¹It goes up to pieces. Therefore, if the enormous number of molecules form a molecule that represents a solution by interaction, a solution can be obtained with a much larger number of parallels than the parallel calculation performed using a conventional computer. In addition, since this reaction can be performed even in a solution of less than 1 mL, even if the solution is heated and cooled, almost no energy is consumed. Computers using these DNAs are expected to exceed the processing speed of Neumann-type computers when applied to large multivariable problems. However, at present, no device has been developed that can withstand practical use and that can effectively perform molecular calculations using DNA molecules.
[0009]
[Problems to be solved by the invention]
In view of the above situation, an object of the present invention is to basically provide a method capable of executing calculations faster than a conventional electronic computer by utilizing the parallelism of molecular operations. is there.
[0010]
[Means for Solving the Problems]
The above object is achieved by, for example, the following means. That is,
A molecular computer that performs molecular calculation by a chemical reaction of the molecule,
A molecular control unit comprising: a reaction control unit that controls a reaction of a chemical reaction of a molecule; and an automatic detection unit that generates a chemical reaction of the molecule based on the reaction control of the reaction control unit and detects a reaction result of the chemical reaction. When,
A variable and a constant defined in association with a coding molecule obtained by encoding a molecule based on a nucleic acid sequence, and a program described by a function defined in association with the operation of the coding molecule, the operation of the coding molecule A control command generating unit that converts the control command into a control command for driving the reaction control unit that performs a reaction control of the chemical reaction of the corresponding molecule and generates a procedure of the control command.
An electronic calculation unit having an output unit for outputting a procedure of the control instruction
It is a molecular computer equipped with:
[0011]
Here, the reaction control includes control of an operation that causes a reaction (such as a PCR reaction) and control of an additional operation that causes a reaction (such as mixing). With this method, it is possible to perform parallel computation faster than a conventional electronic computer.
[0012]
Further, with such a device, for example, it is also possible to determine the genotype and the state of gene expression.
[0013]
Such a device is useful as a molecular computer for performing massively parallel computation at high speed, which is advantageous for solving difficult mathematical problems such as the NP complete problem. Of course, it can also be used for gene analysis.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
In one aspect of the present invention, a computer for executing an information processing method using a nucleic acid is disclosed.
[0015]
As used herein, "nucleic acid" refers to all DNAs and RNAs including cDNA, genomic DNA, synthetic DNA, mRNA, total RNA, hnRNA and synthetic RNA, as well as peptide nucleic acids, morpholino nucleic acids, methylphosphonate nucleic acids and S -Including artificially synthesized nucleic acids such as oligonucleic acids. Also, herein, "nucleic acid", "nucleic acid molecule" and "molecule" may be used interchangeably.
[0016]
Hereinafter, the present invention will be described using examples.
[0017]
I. Molecular computer
1. Overview
According to a preferred embodiment of the present invention, there is provided a computing device for performing an information processing method using a nucleic acid. Such a computing device is a molecular computing device including an electronic computing unit and a molecular computing unit, wherein the electronic computing unit substantially controls the function of the molecular computing unit, and performs molecular control based on the control. It is a molecular computer on which the operation is performed. The molecular operation section is a section that substantially executes calculations by executing chemical reactions of molecules.
[0018]
The calculation method underlying the calculation device of the present invention is such that the molecular calculation unit performs the calculation using the parallelism of the molecular calculation described above, and the electronic calculation unit displays the calculation result based on the obtained data. How to The present invention discloses a calculation device that performs parallel processing of data and gene analysis efficiently by performing calculations using nucleic acid molecules. In addition, by expressing data and programs using nucleic acid molecules and replacing the operations defined by the programs with molecular reactions and executing reactions, the memory capacity and order of parallel processing are orders of magnitude higher than conventional electronic computers. To achieve.
[0019]
In developing the present calculation device, the present inventors first change the contents of a calculation program when executing an operation using a molecular computer into a data format that can be recognized and executed by the molecular computer. It is noted that is preferred. Specifically, before the calculation is performed by the molecular operation unit, the molecule is converted into a code molecule that is previously linked to a specific code. Then, data such as variables and constants in the calculation program are automatically converted into code molecules using the conversion rules. Furthermore, it has been found that performing such a conversion operation in the electronic calculator is preferable for improving the calculation speed of the entire calculation device.
[0020]
In other words, the molecular computer of the present invention is a computer in which the molecular computer and the electronic computer share functions complementarily to each other. Therefore, it is easy to use, and it is also possible to solve the NP complete problem at a high speed and directly analyze information on biomolecules such as nucleic acid molecules. That is, the computing device of the present invention uses the electronic computing unit to cover disadvantages of the molecular computing unit such as slow character input and display and simple four arithmetic operations. As a result, the calculation speed of the entire calculation device can be improved. Therefore, the molecular computing device of the present invention is a hybrid molecular computing device in which the functions that the electronic computer can realize at high speed are handled by the electronic computing unit, and the other parts are handled by the molecular computing unit.
[0021]
FIG. 1 shows a basic computing device of the present invention. The computer of the present invention is roughly divided into an electronic computer 21 and a molecular computer 22. The input section 11 for inputting data and the output section 20 for outputting data are connected to the electronic calculation section 21. The input section 18 and the output section 19 for inputting data are connected to the molecular operation section 22.
[0022]
The electronic calculator 21 is realized by a general electronic computer. Specifically, it includes a calculation unit 14, a storage unit 13, an input / output control unit 12, and a communication unit 23. The storage unit 13, the input / output control unit 12, and the communication unit 23 are connected to the arithmetic unit 14. The input / output control unit 12 is connected to the input unit 11 and the output unit 20. Further, it may include other components of a general electronic computer. The calculation unit 14 is realized by, for example, a CPU. The storage unit 13 is realized by a semiconductor memory or the like.
[0023]
The molecular operation unit 22 includes an operation unit 15, a storage unit 16, an input / output control unit 17, and a communication unit 24. The calculation unit 15 and the storage unit 16, the input / output control unit 17, the input unit 18, and the output unit 19 are realized by a reaction device such as a molecule, a container, and a pipettor. That is, it is realized by a device for performing a chemical reaction of molecules, and the device configuration does not correspond to the arithmetic unit 15, the storage unit 16, the input / output control unit 17, and the like in a 1: 1 manner. The configurations of the arithmetic unit 15, the storage unit 16, the input / output control unit 17, and the like show an example of the configuration in a case where reaction devices such as molecules, containers, and pipettes of the molecular operation unit 22 are regarded as the same concept as the electronic operation unit 21. ing. The calculation unit 15, the storage unit 16, and the input / output control unit 17 are connected to the communication unit 24. The input / output control unit 17 is connected to the input unit 18 and the output unit 19. With these molecules and reaction devices, calculation, storage of data, and input / output control of molecular data are simultaneously realized.
[0024]
In the computer of the present invention, the electronic computer 21 controls the molecular computer 22. When a desired program is input to the electronic computer 21, a plan for molecular calculation to be executed in the molecular computer 22 is generated based on the input information. Further, a molecule required for the target molecular calculation is designed, and information obtained thereby is transmitted to the molecular calculation unit 22. The molecular operation unit 22 performs a molecular operation based on the obtained information. For example, the planning of the molecular calculation may be performed by the operation unit 14 searching a table in which the input information stored in the storage unit 13 is associated with the designed molecular calculation, and selecting the corresponding calculation. . The design of the molecule may be performed by the operation unit 14 searching a table in which the input information stored in the storage unit 13 is associated with the molecule to be designed, and selecting the corresponding molecule.
[0025]
As the input unit 11 on the side of the electronic calculation unit 21, for example, any one of input means such as a keyboard and a mouse to which information can be generally input by a person can be used. The output unit 20 on the side of the electronic calculation unit 21 can use any output means such as a display and a printer that can output information in general.
[0026]
In order to realize the operations of these computing devices, the operation unit 14 functions as, for example, a translation / calculation planning / execution unit 14a, a nucleic acid sequence calculation unit 14b, a result analysis unit 14c, and the like. These means 14a to 14c are executed, for example, by reading a translation / calculation planning / execution program, a nucleic acid sequence calculation program, and a detection result analysis program from the storage unit 13. Further, a program for causing the arithmetic unit 14 to execute various functions exists as a program product recorded on a recording medium, is read by a recording medium reading device (not shown), is stored in the storage unit 13, and is stored in the storage unit 13. May be executed. In addition to functioning as these units 14a to 14c, the calculation unit 14 also functions as a unit that a normal computer executes.
[0027]
Further, the output of the calculation result obtained in the molecular operation unit 22 may be transmitted to the electronic operation unit 21 via the communication units 24 and 23, and may be subjected to desired data processing by the operation unit 14 or the like. For example, the calculation results are totaled and calculated in a format corresponding to the processing format stored in the storage unit 13 in advance. The result of the arithmetic processing is finally output to the output unit 19 or the output unit 20 in a desired form. During these processes, the states of the operation unit 15 and the storage unit 16 of the molecular operation unit 22 are indirectly grasped as information received through the communication units 23 and 24 in the electronic operation unit 21.
[0028]
On the other hand, the molecular operation unit 22 is a unit that performs molecular calculation by experimentally synthesizing a nucleic acid molecule or performing a desired operation reaction using the synthesized nucleic acid molecule. An example of the molecular calculation performed in the molecular calculator 22 will be described later.
[0029]
The calculation unit 14 of the electronic calculation unit 21 converts the initial value input from the input unit 11 into a code molecule representation in addition to the processing performed by the calculation unit of a normal computer, and corresponds to the procedure or function of the calculation program. To perform the operation of the code molecule to be performed, and to create the execution procedure of the operation from the stored calculation program. The operations performed by the electronic calculation unit 21 include, for example, assignment of molecules to variables, assignment of reaction solution storage containers for reagents and the like, reaction equipment (for example, a pipette tip, etc.) and reaction solution during molecule calculation. Allocation of the equipment, allocation of the movement and operation of the reaction equipment such as the container and the pipettor, setting of the temperature control of the thermal cycler, and determination of the execution sequence of the calculation.
[0030]
The basic processing of the computing device will be described according to the processing flow of FIG.
[0031]
Note that among the following (S1) to (S9), the steps of (S1), (S2), (S3), (S5), (S8), and (S9) are executed by the electronic calculator 21. (S4) is executed by the nucleic acid synthesizer 30. (S6) and (S7) are executed by the molecular operation unit 22.
[0032]
(S1) A desired calculation program (original program) is input from the input unit 11 of the electronic calculation unit 21, and information such as the input calculation program, constants and variables of the calculation program is stored in the storage unit 13.
[0033]
(S2) This is a program translation process, and is composed of (S2a) to (S2c). The source program stored in S1 is first sequentially converted into a program (S2a). Then, the data (variables and constants) included in the sequential program is converted into a code molecule representation by the arithmetic unit 14 according to the corresponding data stored in the storage unit 13 in advance. On the other hand, the procedures or functions included in the sequential program are converted into the operation reactions of the corresponding code molecules (S2b). By this (S2b), a calculation procedure table expressing the processing procedure by the calculation reaction is generated. Next, the operation procedure is converted into an execution procedure, and an execution procedure table is generated (S2c). The translation of the program in (S2) is also referred to as “creation of a calculation reaction procedure”.
[0034]
(S3) Calculation processing of nucleic acid sequence. Specifically, this calculation process is a simulation for obtaining a molecule that needs to be newly generated among the code molecules included in the operation procedure table obtained in (S2b). This process need not be performed if a previously designed nucleic acid sequence is used.
[0035]
(S4) Nucleic acid synthesis processing. The calculation result obtained in (S3) is sent to the nucleic acid synthesizer 30, and a coding molecule, which is a nucleic acid having each sequence designed in (S3), is synthesized. This synthesis is not necessary if a nucleic acid designed and synthesized in advance is used. When the synthesis is not performed, the synthesized nucleic acid is input from the input unit 18.
[0036]
(S5) The execution procedure in the execution procedure table is converted into a control instruction for giving an instruction to the molecular operation unit 22, and a control instruction procedure table is generated. This control instruction procedure table is output to the molecular operation unit 22.
[0037]
(S6) In accordance with the control command procedure input in (S5), the molecular operation unit 22 executes a chemical reaction of a molecule corresponding to the operation reaction of the program using the operation molecule prepared in S4. Thereby, an automatic calculation using the molecule is performed.
[0038]
(S7) Detection and identification of a reaction product generated by the chemical reaction performed in (S6) are performed.
[0039]
(S8) The detection result obtained in (S7) is analyzed.
[0040]
(S9) The final data calculated in S7 is processed so as to match the output format stored in the storage unit 13 in advance, and output from the output unit 20. From the output result, the calculation result of the program can be confirmed.
[0041]
Such a process can also form a loop if desired. For example, a result obtained in each step is compared with a condition or the like stored in the storage unit 13 in advance, and depending on a result of the comparison, any one of the previous steps is performed alone or in combination with a plurality of steps. It is also possible to set so that this is repeated.
[0042]
FIG. 5 shows an example of a processing flow when the loop processing is formed. As shown in FIG. 5, the branching process is determined based on the analysis result. By this determination, the program to be processed after the branch processing is determined. Then, returning to (S2a), the determined program is automatically executed again by the molecular operation unit (S6) after conversion to sequential processing and the like.
[0043]
In the case of FIG. 5, it is necessary to generate a control instruction procedure table from the level of the source program for each branch process. Alternatively, the processing flow shown in FIG. 6 may be used.
[0044]
FIG. 6 shows a case where branch processing in the source program and all control instruction procedures after the branch processing have been generated in advance. In this case, if only the determination result of the branch processing is obtained in (S10), the process proceeds to (S12), and it is only necessary to determine the next control instruction procedure in the control instruction procedure table generated in advance (S6). Thereby, the process of translating the program every time the branch process is performed becomes unnecessary.
[0045]
The application of the processing flow as shown in FIGS. 5 and 6 is not limited to the case of the branch processing. In the processing, the calculation result in the molecular calculation unit 22 is once fed back to the electronic calculation unit 21, and the present invention can be applied to all processing when the procedure is changed based on the calculation result.
[0046]
In addition, when there are steps that are difficult to automate, or when automation requires a large scale device, for example, when performing an experimental operation that is preferably performed manually such as culturing in cloning or colony picking, the nucleic acid is output to the output unit 19 once. The experiment operation to be performed is displayed on the output unit 20. Thereafter, the operator may manually operate the reaction device in the molecular operation unit 22 and then input the nucleic acid to the input unit 18 again and input a restart from the input unit 11, or the input may be automatically detected. Then, the calculation may be started again. In this way, the processing sharing between the human and the computer can be changed so that a part of the processing is executed by a human.
[0047]
In this step, if the calculation program aims to obtain a desired molecule and it is not necessary to display the calculation result, the desired molecule (reaction product) obtained in (S6) may be used as a final output. In this case, in the flow of FIG. 4, steps after (S7) are omitted. For example, this is the case where a calculation for selecting an oligo DNA that specifically detects a specific gene is performed. In this case, the molecule obtained as a result of the detection performed by the molecular operation unit 22 is directly output from the output unit 19.
[0048]
In addition, after the calculation result is not converted to the description level of the source program, the reaction product is detected by the molecular calculation unit 22, and the end of the calculation is output to the output unit 20 in (S9) without executing (S8). May be. In this case, the flow is as shown in FIG.
[0049]
The conversion into the code molecule in the step (S2) can be performed by reading a molecule conversion table recorded in advance in the storage means, searching for data included therein, and reading the corresponding data. is there. The molecule conversion table may be, for example, a table in which code molecules used for molecular calculation and information input to the electronic calculation unit 21 are associated with each other.
[0050]
Further, in the conversion of the procedure of the calculation program or the function into the operation reaction in the step (S2), the procedure conversion table recorded in the storage means is read in advance, and the corresponding data is read by referring to the table. Can be performed. The procedure conversion table may be, for example, a table in which the steps of the molecular calculation actually performed, the order in which the molecular calculation is performed, the number of repetitions, and the like are associated with the information input to the electronic calculation unit 21.
[0051]
Further, the information obtained in each of the above steps may be stored in the storage unit 13 for each step, or may be stored in at least a part of the steps.
[0052]
In the above-described procedure, it has been described that the synthesis of nucleic acid molecules is performed in the operation unit 15 provided in the molecular operation unit 22. Synthesis of molecules may be performed. In this case, the molecular synthesis unit is not included in the molecular operation unit 22 and the electronic operation unit 21, but is connected to the electronic operation unit 21 and the molecular operation unit 22. Further, as described above, when all the encoding molecules necessary for performing the molecular calculation are already prepared in the molecular operation unit 22, the nucleic acid sequence calculation (S3) and the nucleic acid synthesis (S4) are unnecessary. FIG. 8 shows a flow in this case.
[0053]
The operation unit 15 in the molecular operation unit 22 includes the following units required for performing an operation reaction using nucleic acid molecules.
[0054]
There, a reaction section for performing various reactions, a nucleic acid holding section for holding nucleic acids necessary for the reaction, a reagent holding section for holding reagents and buffers required for the reaction, and an enzyme for the reactions Enzyme holding section, heating means for heating each part as desired, cooling means for cooling each part as desired, pipetting means such as pipetting pipette, pipette and reaction vessel It is possible to equip all or some of them with washing means for washing and control means for controlling various operations.
[0055]
The calculation unit 15 may include a detection unit for detecting and identifying a reaction product generated by a target calculation reaction. However, the detection unit does not necessarily need to be included in the calculation unit 15. This is because the detection unit may be a very large device depending on the detection means used. The detection unit may be arranged so as not to be included in the electronic operation unit 21 and the molecular operation unit 22 but to be connected to the electronic operation unit 21 and the molecular operation unit 22.
[0056]
For example, the detection in the present apparatus may be performed by electrophoresis of a reaction product generated by an arithmetic reaction. In this case, the length of the nucleic acid molecule is determined by detecting the position of a peak obtained by electrophoresis, and the determination is performed. Alternatively, the reaction product may be supplied to a DNA sequencer, and the sequence may be detected based on data output from the DNA sequencer to determine which of the encoding nucleic acids was initially assigned in the experiment. Alternatively, the method may be implemented by equipping a DNA microarray and a scanner to read the sequence of the encoded nucleic acid by a hybridization method.
[0057]
The automatic control of the molecular operation unit 22 based on the control command in (S6) can be manually performed. FIG. 9 shows a flow in this case. As shown in FIG. 9, (S61) is provided instead of (S6). In this step (S61), the operator manually operates each device of the molecular operation unit 22 with reference to the control instruction procedure table generated in (S5) and printed in advance with a control instruction procedure table by a printer or the like. As described above, even when the operator operates manually, since the control procedure in the molecular operation unit 22 is clearly shown, the computation efficiency is much more efficient than when the conventional molecular operation unit 22 is used. Processing becomes possible.
[0058]
The output can be obtained from the reaction result itself, that is, the nucleic acid, without detecting the reaction result of the molecular operation unit 22 by the detection unit. In this case, steps after (S61) in FIG. 9 are omitted.
[0059]
In order to create a series of calculation procedures, that is, a sequence for performing calculations, in the electronic calculation unit 21, an execution sequence for calculation is uniquely determined from an input program, a problem, and its initial value. preferable. For example, in the case of a 3SAT program, the loop of the program is unrolled, and the conditional branch is determined in advance by referring to the problem. Thereby, a calculation plan by the translation / calculation plan drafting / executing unit 14a can be determined. Therefore, all calculation plans can be determined at the time when the calculation of the program starts. However, depending on the program, there may be a case where it is necessary to insert an array detection operation in the middle of the loop to check the array and then execute a conditional branch. In such a case, the detection operation may be performed, and then the calculation may be performed so as to make a calculation plan for the next stage according to the result.
[0060]
Further, in the computing device of the present invention, for example, it is possible to convert cDNA synthesis into a function and store the function in the storage unit 13 of the electronic computing unit 21 in association with the operation in the molecular computing unit 22 for computation. is there. This allows for automatic procedure planning. In addition, by automatically performing an encoding reaction, each processing, for example, a logical operation and / or a sequence search can be automatically performed in parallel.
[0061]
According to the present invention, various types of work robot systems for conventional experiments, clinical uses, and manufacture can be applied to the molecular operation unit 22. In this case, the operation can be performed so as to correspond to the work plan in which the purpose is given the highest priority, without being restricted by the program of the electronic calculation unit prepared in advance.
[0062]
When a human gene is to be converted, for example, in the present invention, it is preferable to use a partial sequence of an organism that is heterologous and has a low base sequence homology, such as a virus. Further, when simply solving an abiotic problem such as 3SAT, it is also possible to arbitrarily design a sequence such that cross-hybridization does not occur and the Tm is almost equal to each other and use it as a coding sequence. If a large number is required, an orthonormalized array calculated by the electronic calculation unit 21 so as to satisfy a desired condition is used.
[0063]
In addition, the chemical operation in which the molecules in the molecular operation unit 22 are directly input enables the operation desired by the electronic operation unit 21. Therefore, for example, when the present invention is used for gene analysis, it is possible to minimize the experimental error. That is, since the calculation processing is performed in the apparatus in the state of the molecule in correspondence with the nucleic acid to the encoding nucleic acid, an experimental error is suppressed. Further, by using the molecular computer of the present invention, the calculation time and the running cost can be reduced.
[0064]
Furthermore, by expressing the gene analysis by a program, the experiment can be automatically calculated and executed. In addition, for example, the expressed gene is converted into an encoded nucleic acid by performing an encoded OLA (Oligonucleotide Ligation Assay). Then, if PCR amplification is performed using a common sequence at the 3 'and 5' ends of the encoded nucleic acid, the expression ratio thereof may be accurately detected. Details of the encoding OLA will be described later. The conversion to the encoded nucleic acid can be realized by means other than the encoded OLA.
[0065]
2. Detailed description
(1) Specific example of molecular computer
Hereinafter, the present invention will be described in more detail. FIG. 3 shows a specific example of the computing device of the present invention. The computer shown in FIG. 3 includes an electronic computer 21, a molecular computer 22, and a nucleic acid synthesizer 30. The electronic operation unit 21 includes a communication unit 23, and the molecule operation unit 22 includes a communication unit 24. By transmitting and receiving data between the communication unit 23 and the communication unit 24, information communication between the electronic calculation unit 21 and the molecular calculation unit 22 becomes possible. In the example of FIG. 1, the case where the nucleic acid synthesizer is incorporated in the molecular operation unit 22 is shown, but in FIG. 3, it is provided separately from the molecular operation unit 22.
[0066]
The communication unit 23 of the electronic calculation unit 21 is electrically connected to the nucleic acid synthesizer 30 so that the nucleic acid synthesizer 30 can execute nucleic acid synthesis based on the nucleic acid synthesis command transmitted from the communication unit 23. The nucleic acid synthesizer 30 includes, for example, a control unit that controls a nucleic acid container and a nucleic acid molecule based on a command from the communication unit 23. Then, the nucleic acid synthesized under the control of the control unit is transported to the molecular operation unit 22 together with the nucleic acid in the nucleic acid container. Note that the nucleic acid synthesizer 30 may be provided in the molecular operation section 22. In this case, there is no need to provide a transport system for transporting the nucleic acid from the nucleic acid synthesizer 30 to the molecular operation unit 22 separately from the molecular operation unit 22. The transport of the nucleic acid container may be performed by a transport mechanism or manually.
[0067]
Note that the nucleic acid synthesizer 30 does not have to be connected to the electronic calculator 21. In this case, data necessary for nucleic acid synthesis may be output by the electronic calculator 21 and nucleic acid synthesis may be manually performed based on the output result.
[0068]
The electronic calculation unit 21 includes a translation / calculation planning / execution unit 14a, a nucleic acid sequence calculation unit 14b, and a detection result analysis unit 14c as components equivalent to the calculation unit 14 in FIG. The arithmetic unit 14 functions as these parts 14a to 14c by reading a predetermined program from the storage unit 13. The input unit 11, the display unit 20 a as the output unit 20 in FIG. 1, and the printer 20 b are connected to the respective units 14 a to 14 c via the input / output control unit 12.
[0069]
The molecular operation unit 22 includes an automatic control unit 151 functioning as an operation unit 15, a storage unit 16, an input / output control unit 17, an input unit 18, and an output unit 19, an XYZ control pipetter 152, a thermal cycler reaction vessel 153, and beads in FIG. It has a container 154, an enzyme container 155, a buffer solution container 156, a nucleic acid container 157, a detection unit 158, a temperature control means 159, and a transport mechanism 160.
[0070]
The molecular operation unit 22 includes an automatic control unit 151 connected to the communication unit 24. The automatic control unit 151 includes an XYZ control pipetter 152, a thermal cycler reaction container 153 for performing a nucleic acid reaction, a bead container 154 for holding beads, an enzyme container 155 for holding an enzyme, and a buffer for holding a buffer. In order to automatically control the liquid container 156, the nucleic acid container 157 holding the nucleic acid, the temperature control means 159, and the transport mechanism 160, a drive signal is output to each of these components (152 to 157, 159, 160). Each of these components (152 to 157, 159, 160) receives the drive signal and is driven to control the reaction of molecules and reaction conditions. Therefore, each of these components (152 to 157, 159, 160) functions as a reaction control unit as a whole, and each functions as a reaction control element.
[0071]
The control by the automatic control unit 151 is executed based on a control command transmitted from the electronic calculation unit 21 via the communication unit 24. The reaction vessels 153 to 157 may be any vessels, such as a beaker, a test tube or a microtube, as long as they contain a solution containing a nucleic acid or the like.
[0072]
The thermal cycler reaction vessel 153 may be a reaction vessel whose temperature can be adjusted. For example, it can be realized by combining with a generally used thermal cycler. Further, in this example, an example in which beads are used as the carrier has been described, but other carriers can be used. In that case, appropriate components can be provided instead of or in addition to the bead container 154. The enzyme container 155 may be provided with a temperature control means 159 such as a cooler to protect the enzyme from deactivation. If necessary, a temperature control means 159 such as a cooler or a heater may be arranged in another holding means.
[0073]
The reaction in the thermal cycler reaction vessel 153 is executed according to a control command. For example, after the contents of a desired volume are collected from the various containers 154 to 157 by the XYZ control pipetter 152, the reaction is carried out in accordance with conditions such as a desired temperature. The movement in each part is controlled by the automatic control unit 151 in the molecular operation unit. The XYZ control pipetter 152 is a pipetter controlled by the automatic control unit 151 to move in the XYZ directions and / or up and down as desired.
[0074]
The transport mechanism 160 transports, for example, each of the containers 153 to 157 to or from another container accommodating unit (not shown) or to the detecting unit 158 based on a control command from the automatic control unit 151.
[0075]
After completion of the reaction, the detection unit 158 detects and identifies the reaction product. The detection unit 158 may be realized by any detection means or analysis means generally used for detecting and analyzing a nucleic acid molecule, such as an electrophoresis apparatus, a sequencer, a chemiluminescence photometer, and a fluorometer.
[0076]
The control command generated by the electronic calculator 21 is output to the communication unit 23. The communication unit 23 transmits these control commands to the molecular operation unit 22 via the communication unit 24. The molecular operation unit 22 controls each component based on the received control command to execute the molecular operation. The calculation result is transmitted by the communication unit 24 to the electronic calculation unit 21 via the communication unit 23.
[0077]
The communication units 23 and 24 may be realized by wired communication means such as cables, or may be realized by wireless communication means such as radio waves.
[0078]
The nucleic acid synthesis instruction generated by the electronic calculation unit 21 is output to the communication unit 23. The communication unit 23 outputs these nucleic acid synthesis instructions to the nucleic acid synthesizer 30. The nucleic acid synthesizer 30 controls each component based on the received nucleic acid synthesis command to execute nucleic acid synthesis. The nucleic acid container containing the nucleic acid molecule as a result of the synthesis is transported to the molecular operation unit 22. Of course, only the nucleic acid molecules in the nucleic acid container may be carried into the nucleic acid container of the molecular operation unit 22.
[0079]
An example of the function of the calculation unit 14 of the electronic calculation unit 21 will be described below with reference to the flowchart of FIG.
[0080]
(Translation / calculation planning / execution unit 14a)
The translation / calculation planning / executing unit 14a translates a source program including the initial values input from the input unit 11 to generate an execution procedure table ((S2a) to (S2c)). Control instruction generation processing (S5) for converting a calculation plan level (control instruction) to generate a control instruction procedure table, and calculation plan execution processing for causing the molecular calculation unit 22 to execute the calculation plan based on the generated calculation plan. Execute two processes.
[0081]
The translation process includes a first conversion process to a third conversion process.
[0082]
The first conversion process (S2a) is a process of converting a source program including an algorithm such as a branching process or a loop process into a sequential processing program mainly including a sequential processing algorithm that can be converted into a control instruction.
[0083]
The second conversion process (S2b) is a process of converting each command of the sequential processing program into a chemical reaction level expression using a code molecule and a calculation reaction of the code molecule to generate a calculation reaction procedure table.
[0084]
The third conversion process (S2c) is a process of converting the operation reaction procedure table into a driving operation expression of an operation level expression of each component of the molecular operation unit 22 to generate an execution procedure table.
[0085]
The first conversion process may be executed after the second and third conversion processes are executed first without going through the first conversion process.
[0086]
In the second conversion process, data such as constants and variables defined in the source program are associated with code molecules, and procedures or functions defined in the source program are associated with operation reactions. The association between the data and the encoding molecule uses a encoding molecule reference table stored in the storage unit 13 in advance, and the association between the procedure or function and the operation reaction uses a procedure or function-operation reaction correspondence data table.
[0087]
In the code molecule lookup table, the correlation between a certain code molecule and another code molecule is defined. For example, if it is determined in the second conversion process that a certain numerator is associated with a certain variable in the program, the association between the other variables and the numerator is determined by referring to the numerator reference table. This is performed by determining an appropriate code molecule based on the correlation with the molecule. The correlation is, for example, the complementarity between a certain coding molecule and another coding molecule. Further, each code molecule is specified up to the level of the base sequence.
[0088]
The procedure or function-operation reaction correspondence data table is a table in which a plurality of control instructions are associated with a procedure. FIG. 11 is a diagram showing an example of a procedure or function-operation reaction conversion data table. In FIG. 11, the operation reaction is represented by a sentence, but the data format actually stored as a data table on the electronic calculation unit 21 is divided into elements such as operations, instruments, arrays, and code molecules. Express in an identifiable manner.
[0089]
The sequential processing program is defined by a combination of data and a procedure or a function. Accordingly, an operation reaction procedure table is generated by assigning a code molecule to a code molecule which is one of the elements in a procedure in which a procedure or function is replaced with an operation reaction. Further, not only the assignment of the coding molecule but also the assignment of the device and the arrangement may be performed in the same manner as the assignment of the coding molecule.
[0090]
FIG. 12 shows an example of the generated operation reaction procedure table. As shown in FIG. 12, an operation reaction and variables and constants for specifying the operation reaction are shown for each operation procedure. In addition, in this calculation procedure, the association with each instrument of the molecular operation unit 22 (for example, the second of the five nucleic acid containers 157 having five test tubes T) is not performed. In addition, an operation (for example, extraction) for executing an operation reaction is specified, but the expression form of the operation is not yet at a level that the automatic control unit 151 of the molecular operation unit 22 can recognize as an instruction for controlling each component. .
[0091]
In the second conversion process, data (variables, constants) and code molecules are assigned using the code molecule reference table, but the code molecule reference table actually stores the reagent in the molecule calculation unit 22. No code molecule other than the code molecule prepared as is registered. Therefore, it is necessary to perform nucleic acid synthesis on the unregistered coding molecule to obtain the required coding molecule. Therefore, the translation / calculation plan drafting / executing unit 14a issues a nucleic acid sequence acquisition instruction for the unregistered coding molecule to the nucleic acid sequence calculating unit 14b. The detailed function of the nucleic acid sequence calculator 14b will be described later.
[0092]
In the third conversion process, the operation reaction procedure table is converted to the operation operation of the operation level expression of each component of the molecular operation unit 22 using the operation reaction-execution operation conversion table to generate the execution procedure table. In the operation reaction-execution operation conversion table, a plurality of execution operations are usually associated with one operation reaction. Usually, this is because a plurality of operations for controlling the configuration of the molecular operation unit 22 are required to execute a certain operation reaction. Of course, one execution operation may be associated with one operation reaction.
[0093]
FIG. 13 shows an example of the execution procedure table converted by using the operation reaction-execution operation conversion table. As shown in FIG. 13, a certain calculation procedure is associated with a plurality of execution operations. The content of the procedure is at a level that the automatic control unit 151 of the molecular operation unit 22 can recognize as an instruction for controlling each configuration, but is still associated with the actual configuration of the molecular operation unit 22 (for example, the test tube T is The second of the five nucleic acid containers 157, etc.).
[0094]
The calculation plan generation processing generates a control instruction procedure table based on the execution procedure table obtained by the translation processing. Device data and device constraint data are used to generate the control instruction procedure table.
[0095]
In the calculation plan generation process, first, each execution procedure in the execution procedure table is converted into a control command with reference to the device data. The control command is a command given to the automatic control unit 151. In response to the control command, the automatic control unit 151 controls each device (the pipetter 152, the containers 153 to 157, the transport mechanism 160, and the like in FIG. 3) of the molecular operation unit 22.
[0096]
More specifically, in order to convert to a control command, the device of each execution procedure is assigned to an actual device configuration based on device data. With this assignment, each execution operation specifies an actual device of the molecular operation unit 22 and becomes a level of an instruction capable of controlling the device. In the execution operation of the execution operation table, the automatic control unit 151 cannot recognize the physical configuration. Therefore, for example, when two pipetters 152 are provided in the molecular operation unit 22, it is not known which pipetter 152 is to be operated. On the other hand, the control command is specified up to the level at which the pipette 152a of the two pipetters 152a and 152b is operated.
[0097]
Each time an execution operation is converted into a control instruction, an additional control instruction based on the device constraint is added by referring to the device constraint data. The apparatus constraint condition data includes, for example, the number and capacity of each vessel 153 to 157 such as a reaction vessel, the number and extraction amount of the XYZ control pipetter 152, the number and arrangement of reagents, the temperature control speed by the temperature control means 159, and the transfer mechanism 160 The transport speed, the distance between the components, and the like. The conversion of the execution operation into the control instruction is performed on the assumption that the device resources are infinite. However, if the execution operation is converted into a control instruction on a one-to-one basis, processing that cannot be continued may occur due to restrictions on the device of the molecular operation unit 22. For example, since the number of reaction vessels is two, processing using three reaction vessels at the same time cannot be continued. In this case, such a device constraint condition is registered in the storage unit 13 in advance, and by referring to this, it is possible to set a measure in a case where the process cannot be continued. As an example of the treatment, for example, a control command for issuing a warning requesting the operator to add a reaction vessel to the operator by a notifying means such as an alarm (not shown) is added, and whether or not the reaction vessel has been added is detected by a detector (not shown). Control instructions. A control instruction number is added for each additional control instruction based on the device constraint. It is preferable that an operation procedure number, an execution procedure number, and the like are associated with the control instruction number. The obtained control command is associated with the control target because the control target in the molecular operation unit 22 is clear.
[0098]
FIG. 14 shows an example of the control instruction procedure table thus obtained. As shown in FIG. 14, the control target is clear and the sequence takes into account the device constraints.
[0099]
If a series of control commands are standardized, such as a control command for aspirating a nucleic acid solution by an XYZ control pipettor and a control command for transferring to a reaction vessel, the procedure is converted into a control command set. The control instruction set may be transmitted to the molecular operation unit 22. In particular, a plurality of control commands generated from one operation reaction can be easily standardized. In this case, it is preferable that the automatic control unit 151 has a data table in which a control instruction set and a control instruction are associated with each other, and converts the control instruction set into a control instruction based on the data table, and controls each component. Further, the automatic control unit 151 may not have such a data table, and each device (for example, the XYZ control pipetter 152, the transport mechanism 160, etc.) may automatically execute the control instruction set.
[0100]
In the calculation plan execution processing, the control instructions are output to the automatic control unit 151 in the order of the control instruction numbers based on the control instruction procedure table obtained in the calculation plan generation processing. Thereby, the automatic control unit 151 can automatically control each device necessary for molecular calculation. Of course, if manual intervention is required based on the restrictions of the device, such control is also performed on the device. For example, if a warning for refilling the reaction container at the designated position is alarmed, the operator refills the designated reaction container at the designated position. This eliminates the need for the operator to constantly monitor the progress of the calculations performed by the molecular calculation unit 22.
[0101]
In the above-described processing example, a case has been described in which a control instruction is generated in consideration of device restrictions based on device restriction condition data in the control instruction generation processing. However, this processing may be omitted when there is sufficient device resources.
[0102]
Further, the case has been described in which the control instruction procedure table is generated in advance and the molecular calculation is executed after specifying the control target, but the present invention is not limited to this. For example, each device (XYZ control pipetter 152, containers 153 to 157, etc.) is provided with a use presence / absence check sensor for checking the use of device resources each time each control command is executed, and an automatic control unit is provided based on the detection data of the sensor. The system 151 may automatically assign control targets.
[0103]
The device data is not limited to the configuration as shown in FIG. 3 and can be variously changed according to the device configuration of the molecular computer to be used. As long as the device configuration of the molecular computer to be used and the correspondence between the execution operation and the operation reaction in the device (operation reaction-execution operation conversion table) are registered as device data, any molecular computer of any device configuration can be used. It can be a target of the molecular calculation of the present invention. That is, by registering the device data, the device constraint data, and the operation reaction-execution operation conversion table for each different molecular operation unit, the molecular operation unit having a different device configuration can be easily changed without changing the other configuration of the electronic operation unit. Perform molecular calculations.
[0104]
Further, the first to third conversion processes do not necessarily need to be performed by the electronic calculator 21. By making the automatic control unit 151 of the molecular operation unit 22 function in the same manner as the operation unit 14 of the electronic operation unit 21, the automatic control unit 151 may be executed.
[0105]
Further, the translation / calculation plan drafting / executing unit 14a described above has been described with reference to the case where the program is sequentially converted into the processing format to create the control instruction procedure, but the present invention is not limited to this. For example, when there is a conditional branching process based on a certain calculation result, the calculation result (detection result) is temporarily received from the detection unit 158 by the translation / calculation plan drafting / execution unit 14a. Then, a conditional branch process is performed based on the calculation result, and the control instruction procedure after the branch is newly given to the molecular operation unit 22. Thereby, even when the conditional branch occurs, the molecular calculation can be automatically continued. In addition, automatic calculation can be performed by the electronic calculation unit 21 once receiving the detection result from the detection unit 158 before the control command procedure is completed, and continuing the molecular calculation in which the detection result is reflected.
[0106]
Further, the data-code numerator associating process in the second conversion process is described by referring to the code numerator reference table and automatically executed by the translation / calculation plan drafting / executing unit 14a. However, the present invention is not limited to this. Absent. Variables and constants and code molecules may be manually set.
[0107]
(Nucleic acid sequence calculation unit 14b)
The nucleic acid sequence calculation unit 14b calculates a nucleic acid sequence corresponding to the code molecule given in the data-code molecule association process in the translation process in the translation / calculation plan drafting / executing unit 14a. Many code molecules are provided in the data-code molecule association processing. Further, the number of code molecules registered in the code molecule reference table is limited. Therefore, the base sequence of the missing coding molecule is calculated by the nucleic acid sequence calculation unit to design the sequence. More specifically, it has an optimum melting temperature (Tm) of a double strand and a free energy value of a large secondary structure that does not cause cross-hybridization with a coding molecule already registered in the coding molecule lookup table. Design the base sequence. The cross-hybridization is checked by simulating the hybridization with the sequence already in the encoding molecule reference table. Melting temperatures and free energy values can be easily calculated from the array. Therefore, an appropriate array of melting temperatures and free energy values is selected. By these calculations, the sequence of the nucleic acid to be synthesized is determined.
[0108]
The nucleic acid sequence calculator 14b transmits the obtained nucleic acid sequence calculation result to the nucleic acid sequence synthesizer 30 via the communication unit 23. The nucleic acid sequence synthesizer 30 synthesizes the nucleic acid automatically or manually based on the given nucleic acid sequence calculation result, stores the obtained nucleic acid solution in a nucleic acid container, and automatically or manually conveys it to the molecular operation section 22. The nucleic acid sequence synthesizer 30 transmits the nucleic acid container identification information for identifying the nucleic acid container containing the synthesized nucleic acid solution to the translation / calculation plan drafting / executing unit 14a. The translation / calculation plan drafting / executing unit 14a generates a control instruction procedure table in the control instruction generation processing based on the received nucleic acid container identification information. Accordingly, even when a calculation device not registered as device data is introduced into the molecular calculation unit 22, it can be incorporated into automatic molecular calculation.
[0109]
Note that the nucleic acid sequence calculator 14b does not need to be executed by the electronic calculator 21 as long as the nucleic acid solution is prepared by another device. When all the nucleic acid solutions having the required base sequence are registered, the nucleic acid sequence calculation and the nucleic acid synthesis can be omitted.
[0110]
(Detection result analysis unit 14c)
The detection result analysis unit 14c receives the detection result obtained by the detection unit 158 of the molecular operation unit 22 via the communication units 24 and 23, and outputs the result to the display unit 20a or the printer 20a. It is preferable to analyze not only the detection result but also, for example, the detection result and output the analysis result. Further, the results calculated by the translation / calculation planning / executing unit 14a and the nucleic acid sequence calculating unit 14b may be output as appropriate.
[0111]
The detection result from the detection unit 158 is the reaction result of the DNA itself, and is often not the information desired by the user. Therefore, it is necessary to convert the reaction result into information that can be understood by the user. This conversion work can be automatically executed by the detection result analysis unit 14c. This is because in this conversion work, reverse conversion of the conversion work from the source program as information that can be understood by the user to the control instruction procedure table recognizable by the molecular operation unit 22 may be performed.
[0112]
Specifically, the analysis of the detection result refers to, for example, a control command procedure table, device data, an execution procedure table, an operation reaction procedure table, a code molecule reference table, a procedure or function-operation reaction conversion table, etc. from the storage unit 13. It can be performed by doing.
[0113]
Consider a case in which an analysis process for converting a certain reaction result to a solution level of a source program is performed. The reaction result is a chemical reaction, and the detection result is obtained by the detection unit 158. Which control command, execution procedure, operation reaction, and code molecule each data of the detection result of the detection unit 158 corresponds to is a control instruction procedure table, an execution procedure table, an operation reaction procedure table, a code molecule reference table, This is possible by referring to a procedure or a function-operation reaction conversion table. By this analysis processing, the reaction result is converted into information that can be used by the user. Therefore, the user can understand the calculation result without being conscious of the operation content in the electronic calculation unit 22 at all.
[0114]
Note that this analysis processing is not essential. For example, there is a case where the reaction result is information that the user needs as it is. In this case, the detection result from the detection unit 158 is not converted at all, or is analyzed by various data expression formats such as an execution procedure level, a control instruction level, an operation reaction level, and a code molecule level. The information that the person needs is obtained. In particular, in the case of genetic information analysis, analysis processing may be performed up to the level of genetic information.
[0115]
An example of a processing procedure in the electronic calculator will be described.
[0116]
First, a source program including an initial value input from the input unit 11 is converted into an internal code by the input / output control unit 12 and then stored in the storage unit 13. The internal code refers to a code that can be electrically recognized by the calculation unit 14 of the electronic calculation unit 21 and data can be processed. Next, the translation / experimental planning / executing unit 14a converts the source program converted into the internal code into a code molecule and a code molecule operation reaction procedure table, and generates an operation reaction execution table. Subsequently, it is stored in the storage unit 13. Further, if necessary, the nucleic acid sequence calculation unit 14b calculates the coding molecule based on the calculation reaction procedure table. This calculation result is sent to the nucleic acid synthesizer 30. The nucleic acid synthesizer 30 synthesizes a nucleic acid based on the obtained calculation result, and transports the obtained result to the molecular operation unit 22. In addition, the nucleic acid synthesizer 30 transmits information for identifying the obtained nucleic acid solution as an apparatus to the electronic calculator 21 in order to incorporate the obtained result into the automatic calculation.
[0117]
Information handled in these steps may be always displayed on the display unit via the result analysis unit 14c. Alternatively, it may be set so that only desired information is displayed at any time. It may be printed out as needed.
[0118]
In the present invention, it is assumed that the molecular operation unit 22 includes at least steps or means involved from the start to the end of a biological specific reaction such as a hybridization reaction, an enzyme reaction, or an antigen-antibody reaction. Here, the start time of the reaction includes at least the stage at which the reaction pair is found to exhibit significant selectivity at the earliest, and includes the latest immediately before the target such as detection and separation can be achieved. In addition, the end point of the reaction may include at least the earliest time point at which the purpose of detection, separation, and the like can be achieved as early as possible, and may include the time point longer than the elapsed time at which the purpose of detection, separation, and the like can be sufficiently achieved. it can.
[0119]
FIG. 2 shows a more detailed arrangement example of the molecular operation unit 22 of the present invention. For example, the molecular calculation section includes an 8-chip rack 31, a chip rack 0, a chip rack 1, a 33, a 96-well multi-titer plate (hereinafter, referred to as MTP) 2, 34, 96. It is possible to arrange a hole MTP No. 35, a 1.5 mL tube rack 36, a 96-well MTP No. 37, a thermal cycler 96-well MTP 38, and a chip disposal port 39. However, this is not restrictive at all, and various changes can be made as desired.
[0120]
(2) Molecular computation in the molecular computation department
(A) Application to 3SAT
The molecular calculation in the molecular computer according to one embodiment of the present invention, which will be described below, may be each embodiment of the present invention described in the above item I.
[0121]
In the following, an example will be described in which a calculation for solving a satisfiability determination problem (3SAT) of a sum-of-products propositional logical formula, which is a typical NP-complete problem, is performed by applying the molecular computer of the present invention.
[0122]
First, the problem of DNA calculation based on the conventional Edelman-Lipton paradigm will be described. The problem with DNA calculations based on the Edelman-Lipton paradigm is that it is necessary to first generate a pool of DNA molecules that represent all solution candidates. The number of candidate solutions for the NP-complete problem increases exponentially with the number of variables. Therefore, as the size of the problem increases, there is a risk that even a DNA computer having a very large memory cannot generate a complete pool including all solution candidates. This is a serious scaling problem within the Edelman-Lipton paradigm.
[0123]
Consider the case where the Edelman-Lipton paradigm is applied to the above-mentioned 3SAT. Assuming that the number of logical variables is 100, the number of solution candidates is 2¹⁰⁰= 1.3 × 10³⁰Also. Expressing one variable as a 15 base pair DNA molecule requires an unrealistic amount of DNA molecules of at least 2 million tons to produce a complete probe. If one solution candidate is represented by one molecule, there is a high risk of being lost in the middle of a calculation reaction. Therefore, a larger amount of DNA molecules is actually required. With 200 variables, at least about 40 trillion times the Earth's mass would require DNA molecules. It must be said that it is impossible to actually solve the problem.
[0124]
In such a situation, the inventors considered solving an NP-complete problem by executing an algorithm based on dynamic programming on a molecular computer (also referred to as a DNA computer). That is, instead of generating all the solution candidates from the beginning as in the Edelman-Lipton paradigm, the system generates candidate solutions of partial problems and selects and extracts solutions from them. This operation is repeated while sequentially increasing the size of the partial problem, so that a solution to the original problem is finally obtained. In this way, the number of generated solution candidates can be significantly reduced.
[0125]
Here, the molecule design described in the item of (1.1) preparation described later is first converted from a source program into necessary information in a translation / experimental planning / execution unit of an electronic calculation unit. Then, a nucleic acid sequence suitable for desired conditions is designed in the nucleic acid sequence calculation unit. Further, the above-described steps can be performed in the calculation unit of the electronic calculation unit. A nucleic acid is synthesized based on the information obtained in the electronic operation unit, and the molecular operation is executed in the molecular operation unit.
[0126]
(B) 3SAT program
3SAT, which is a typical example of the NP complete problem, can be solved based on a dynamic programming algorithm by performing DNA calculation according to the following program, for example.
[0127]
(Equation 1)

[0128]
Here, the symbol “∧” indicates a logical product, “∨” indicates a logical sum, and “¬” indicates a negation.
[0129]
(Equation 2)

[0130]
Using this apparatus, we solved a 3SAT problem with 4 variables and 10 clauses. Equation 1 shows the problem, and Equation 2 shows the solution of the problem. The logical expression of the problem in Equation 1 is x_1,x₂, X₃, X₄, And ten 3-lateral clauses are connected by AND. The purpose is to find a set of values of variables that satisfy this equation, that is, whether or not there is a solution, and if so, a set of values of the variables.
[0131]
(Equation 3)

[0132]
In order to solve this problem, the program shown in Expression 3 is executed by the present apparatus. The description method conforms to Pascal. Specifically, the program (dna3sat) shown in Expression 3 is input as a source program by the input unit 11 in FIG. The input source program is translated to generate a control instruction procedure table through translation and the like, and the molecular calculation unit 22 executes molecular calculations. Then, through the steps (s2) to (s8), the calculation result shown in Expression 3 is obtained.
[0133]
The function dna3sat is a main function that outputs a solution when a solution exists when a problem is given. The function getavsat is incorporated in the function dna3sat. These functions dna3sat and getuvsat are composed of five basic functions: get, amplify, append, merge, and detect. Among the basic functions, “get”, “amplify”, and “append” are executed by the chemical reactions shown in FIGS. 15 to 18 as procedures or functions of the source program.
[0134]
Next, a chemical reaction associated with each basic function and operations associated with the chemical reaction will be described. First, the get (T, + s) and get (T, -s) functions will be described with reference to FIG.
[0135]
The function get is a function that defines a reaction operation for obtaining a single-stranded oligonucleotide containing a sequence s or an oligonucleotide not containing the sequence from a mixed solution T (tube) of oligonucleotides as a calculation reaction of a coding molecule. is there. Here, s indicates a specific sequence of several or tens of bases. get (T, + s) acquires an oligonucleotide containing an s sequence, and get (T, -s) acquires an oligonucleotide containing no s. T, + s, and -s are variables defined in association with a coding molecule obtained by coding a molecule based on a nucleic acid sequence.
[0136]
First, an oligonucleotide having a sequence complementary to s and labeled with 5'-end biotin is placed in T, and annealed with an oligonucleotide having an s sequence.
[0137]
The hybrid formed by this annealing is captured, for example, by magnetic beads having streptavidin bound to the surface. Thereafter, the substrate is washed with a buffer solution of magnetic beads at a temperature at which the hybrid of the oligonucleotide complementary to the s sequence is not dissociated (ie, cold-washed). As a result, an oligonucleotide having no s sequence can be obtained from the buffer, and get (T, -s) can be executed.
[0138]
The function get (T, + s) may be executed by washing the magnetic beads with a buffer solution of a relatively high temperature at which the hybrid is dissociated after the cold wash. Thereby, an oligonucleotide having an s sequence can be obtained from the buffer solution. That is, two functions can be executed by one operation.
[0139]
Next, the reaction of the function append (T, s, e) will be described with reference to FIG. e is a variable defined in association with the encoded molecule obtained by encoding the molecule based on the nucleic acid sequence.
[0140]
This function is to ligate an oligonucleotide having a sequence s to the 3 ′ end of a single-stranded DNA having an e sequence at its 3 ′ end in a mixture T of oligonucleotides by ligation, and Is a function that defines a reaction operation for obtaining a linked single-stranded oligonucleotide as a calculation reaction of a coding molecule. Here, s and e indicate a specific sequence of several bases or several tens of bases as described above.
[0141]
In the reaction of the function append (T, s, e), the following oligonucleotide is used.
[0142]
The 5 'end of the oligonucleotide having the s sequence is phosphorylated. The linking oligonucleotide is labeled with biotin at the 5 'end, and contains a sequence complementary to s at the 5' end and a sequence complementary to e at the 3 'end adjacent to each other. When these are put into the tube T and the reaction is started, the linking oligonucleotide forms a hybrid with the target oligonucleotide sequence e and the s sequence oligonucleotide in T. This hybridization reaction is performed at a relatively high temperature so that a mismatched hybrid is not formed, and ligation is performed using an enzyme having high activity at a high temperature such as Taq ligase. Thereafter, if the hybrid is captured by streptavidin magnetic beads or the like and washed at a high temperature at which the hybrid is dissociated (that is, hot wash), the oligonucleotide having the s sequence linked to the 3 'end can be recovered.
[0143]
Next, according to FIG.₁,. . . . T_n) Describe the reaction of the function. This amplifier (T, T₁,. . . . T_nThe function reaction is performed by amplifying the oligonucleotide contained in the reaction solution T by a PCR reaction, converting the amplified double-stranded oligonucleotide to a single strand,₁,. . . T_nThis is a reaction of dividing into n reaction solutions. In this embodiment, both ends of the oligonucleotide contained in T are provided with a common amplification sequence, so that all the oligonucleotides in T can be amplified by one set of primers. Among the primers, those having a sequence complementary to the 3 'end of the oligonucleotide to be amplified are labeled with biotin at the 5' end. The T oligonucleotide is amplified by this common primer and captured by streptavidin magnetic beads. If this is washed at a high temperature such as 94 ° C. so that all the two strands are dissociated, the strand on the side originally contained in T can be extracted into the buffer solution. This extraction solution is evenly T₁,. . . T_nMay be divided into n reaction solutions.
[0144]
The function “merge” is a function that defines a reaction operation of combining a plurality of solutions of an argument into one solution as a calculation reaction of a code molecule.
[0145]
Finally, the function detect will be described with reference to FIG. Detection can also be performed by manual reaction. An example of a method of performing the function detect by the present apparatus will be described below. Here, an example is shown in which the detection is performed using the detection means disclosed in Edelman's dissertation (Science, 266, 1021-4), which is graded PCR.
[0146]
In FIG. 18, a nucleic acid molecule showing a solution generated by the present apparatus is a sequence in which sequences in which four variables represent true (ie, T, true) or false (ie, F, false) in order from the 5 ′ end are linked. Has. The sequence of this solution is individually PCR amplified with primers that can detect all possible solutions, also shown in FIG. In the figure, the horizontal line drawn above the sequence name indicates the complementary sequence. If a sequence solution as shown at the beginning of FIG. 18 is obtained, a PCR product having the length of the primer pair and the length shown in the drawing is obtained. Confirmation of the presence and length of these products by gel electrophoresis reveals the sequence of the solution from the primers obtained for the PCR products. Note that the function detect also includes an output from a DNA chip.
[0147]
Table 1 summarizes the above functions.
[0148]
[Table 1]

[0149]
The main function of the program shown in the above equation 3 is dna3sat. The basic function described above and the function getuvsat composed of the basic function are incorporated therein. In this way, it is possible to define a new procedure or function by combining the basic functions, and create a program using this procedure or function. First, the variable names in the program are explained.
[0150]
T is an abbreviation for tube.₂Is x₁, X₂2 is a solution containing oligonucleotides representing all four possible truth sets of the two variables. X₁ ^TIs the variable x₁Is true and is 22 bases in length. X₂ ^FIs the variable x₂Are false, and the length is also 22 bases. Therefore T₂X, one of the oligonucleotides contained in₁ ^TX₂ ^FIs 44 bases long, x₁= 1, x₂= 0 is a single-stranded oligonucleotide representing the assignment 0. j and k are integers indicating the progress of the loop, j indicates which node of the logical expression in question is being calculated, and k indicates what variable is being calculated. u, v, and w indicate each literal in each clause of the logical expression in question. Each literal is u if₁= X₁, V₂= X₂, W₁= ¬x₃Becomes The bar above the sequence name X means the complementary sequence. Also, T / F on the right shoulder of X indicates T and F, and X^TAnd X^FMeans
[0151]
Hereinafter, the program of Expression 3 will be described step by step along the flowchart of FIG. In practice, the program interpreter develops the function, conditional branch, and for loop of this program into a series of experimental operations to make an experimental plan. First, in order to simplify the processing of the theoretical formula in the program, literals are sorted in ascending order of variable subscripts in each section. Also, two variables x₁, X₂Solution T with all possible assignments to₂Be prepared.
[0152]
In the function dna3sat, k = 3 of the first for loop, that is, x₃A loop for determining a value to be assigned to is described. First, T₂Amplify. In this case, after amplification, T₂T with the same oligonucleotide as_w ^T, T_w ^FTo obtain a total of three solution tubes. Next, it enters the j for loop in the k loop. Here, the sufficiency of the j-th node of the logical expression is evaluated. In the first conditional branch, if j = 1, the third literal w of the first clause₁Is x₃Checks for equality. In this case, the electronic calculator 21 checks a logical expression in advance, determines whether or not to execute the then, and determines whether or not to execute the experiment. In the second conditional branch in the loop, the third literal w of the first clause₁Is x₃Therefore, the process after the then is executed.
[0153]
The function getuvsat is defined separately in the lower part. For each argument, T, T_w ^TTube (content and concentration are T₂Same for u), u for u₁(= X₁), V for v₁(= X₂). First T_u ^TIs T_w ^TX from₁ ^TThe oligonucleotide containing is extracted to produce a solution. This gives T_u ^TX₁ ^TX₂ ^T, X₁ ^TX₂ ^F内容Then T '_u ^FIs T_w ^TX from₁ ^TExtract the oligonucleotides that do not contain to generate a solution. This allows T '_u ^FX₁ ^FX₂ ^T, X₁ ^FX₂ ^F内容Third, T_u ^FIs T '_u ^FX from₁ ^FThe oligonucleotide containing is extracted to produce a solution. This gives T_u ^F= ｛X₁ ^FX₂ ^T, X₁ ^FX₂ ^F内容There is a comment that this line here can be ignored, but it is clear from the contents of the resulting solution. Since this line is added so that no error occurs in the experiment, it is executed just in case. If ignored, the previous line T '_u ^FTo T_u ^FRead as 4th T_v ^TIs T_u ^FX from₂ ^TThe oligonucleotide containing is extracted to produce a solution. This gives T_v ^T= ｛X₁ ^FX₂ ^T内容Finally, by the function merge, T_u ^TAnd T_v ^TAre mixed and T^TIs generated and output by return. T^TX₁ ^TX₂ ^T, X₁ ^TX₂ ^F, X₁ ^FX₂ ^T｝ And the solution d in the original dna3sat function_w ^TWill be treated as a name. T^TIs that the first and second literals in section 1 are x₁∨x₂The second literal ¬x₃X is true where the first clause of the first clause is true no matter what value₁, X₂4 shows a set of value assignments to. As described above, the function getuvsat is a function for selecting an oligonucleotide representing an assignment in which a value is assigned to two literals and the assignment is true no matter what is assigned to the variable of the third literal. It is.
[0154]
Next, j is incremented by one again in the k = 3 loop in dna3sat, and when j = 2, the calculation proceeds to the second section. In this section, the third literal is $ x as in section 1.₃Therefore, it is the lower if statement that is effective for conditional branching. Execute getuvsat as in section 1. Note that j = 1 and T_w ^TIs generated. Also, v₂Is v₂= ¬x₂Since this is a literal with negation, T of getuvsat function_v ^TX in the formula for_v ^TIs X₂ ^FAnd execute the function get. In this way, the newly obtained T at j = 2_w ^TIs ｛X₁ ^TX₂ ^T, X₁ ^TX₂ ^F｝
[0155]
Thereafter, the processing is sequentially performed until j = 10. In the middle, for example, when the third literal is k = 4 as in the fifth and subsequent nodes, the value of j is incremented and the next processing of the loop is performed. If the loop for 10 clauses ends at k = 3, T_w ^TThe 3 'end of is X₂X representing true for the oligonucleotide₃ ^TAppend. Also, T_w ^FThe 3 'end in the X₂X that represents false to the oligonucleotide₃ ^FAppend. Thereafter, the solution after the append is merged, and the process proceeds to a loop of k = 4. When the loop of k = 4 ends, the loop is finally exited and T₄Process the tube with the function detect.
[0156]
Since the calculation is performed as described above, if the number of variables is n and the number of clauses is m, the number of executions of each basic command is
(N−2) × (amplify + 2 × append + merge) + mx × (3 × get + merge)
It is. That is, the 3SAT problem can be solved in a time almost proportional to the number of variables and the number of nodes.
[0157]
Also, in the present invention, the detect part does not necessarily mean that the reaction result is measured, and the reaction result is transmitted to the electronic calculation unit so that it can be used, or is changed or extracted to a form that can be used by the user. What is necessary is just to be characterized in the state where it was. In addition, the detect portion only needs to be characterized in a measurable state, and the measurement may be performed by a person. Therefore, in the present invention, the molecular operation unit only needs to carry out the processes up to the end of the reaction and the provision of a product or product that can be used by the user. The user may use the product or product provided by the device of the present invention for various purposes, for example, diagnosis, treatment, drug discovery, academic research, construction of a biological database, decryption of biological information, etc. And can be used most efficiently.
[0158]
【Example】
Molecular computation
The above-described program of Expression 4 was executed using the molecular computer of the present invention. The configuration of the device is described below. 2 shows a configuration of an apparatus that has executed the program. This device was manufactured by modifying the automatic nucleic acid extraction device SX8G of Precision System Science (PSS). This apparatus is equipped with a computer (corresponding to the electronic calculation unit 21) having an OS of Windows 98 (registered trademark) equipped with an Intel Pentium III CPU (registered trademark) for control, and a reagent tank for actually performing an arithmetic reaction. It consists of a reaction tank, a pipetter capable of controlling the position of XYZ, a spare pipettor tip, and a reaction robot (corresponding to the molecular calculation unit 22) including a thermal cycler (PTC-200 manufactured by MJ Research) whose temperature can be controlled by a computer. . FIG. 2 shows the arrangement of the reaction members as viewed from above the molecular operation unit 22.
[0159]
Delivery of the solution is performed between these reaction containers. Extraction of oligonucleotides with streptavidin magnetic beads is performed using a special tip of a pipettor and a permanent magnet that can be moved closer to or away from the tip. However, two operations that could not be performed by the above system, that is, dispensing a small amount of enzyme of about 1 μL, were performed manually, and capillary gel electrophoresis with the function detect was performed by P / ACE-5510 capillary electrophoresis manufactured by Beckman Coulter. Performed by electrophoresis system.
[0160]
The reagents and their initial arrangement are described. The ligase enzyme was retained outside the instrument as it was manually dispensed as described above. Dispensing was performed using a 2 μL pipettor manufactured by Gilson Corporation.
[0161]
The length of the oligonucleotide indicating the value of each variable represented by X is 22 bases. Also, here X₁ ^TX₂ ^FX in order from the 5 'end side₁An oligonucleotide representing that X is true, X₂Is a single-stranded oligonucleotide having an oligonucleotide sequence that is false. Here, when the sequence name is sandwiched between "[]", it means a sequence complementary to the original sequence. Therefore, for example, [X₁ ^TX₂ ^F] Then X₁ ^TX₂ ^F, The actual sequence is X₂ ^FA sequence complementary to₁ ^TThe sequence is complementary to the sequence.
[0162]
T₂Solution (X₁ ^TX₂ ^T, X₁ ^FX₂ ^T, X₁ ^TX₂ ^F, X₁ ^FX₂ ^F5 pmol of each was dissolved in 20 μL of a ligation buffer solution) and retained on MTP35 (ie, a multititer plate 35). Buffer solution for ligation reaction, streptavidin magnetic beads, B & W solution (the solution in which NaCl was added to TE solution to make the concentration of NaCl 1 M is hereinafter referred to as this solution. This solution is prepared by adding biotinylated oligonucleotide to streptavidin magnetic beads. (Used when capturing or performing a hybridization reaction) was retained in the MTP 37 of FIG.
[0163]
In addition, a biotinylated oligonucleotide [bX₁ ^T], [BX₂ ^T], [BX₃ ^T], [BX₄ ^T], [BX₁ ^F], [BX₂ ^F], [BX₃ ^F], [BX₄ ^F], Each of which was dissolved in 20 μL of a B & W solution at 10 pmol each, and an append oligonucleotide pX having a phosphorylated 5 ′ end.₃, PX₃ ^T, PX₃ ^F, PX₄ ^F, PX₄ ^T, 10 μmol each dissolved in 20 μL of a ligation buffer solution, a ligated oligonucleotide labeled with biotin at the 5 ′ end, [bX₂ ^TX₃ ^T], [BX₂ ^FX₃ ^T], [BX₂ ^TX₃ ^F], [BX₂ ^FX₃ ^F], [BX₃ ^TX₄ ^T], [BX₃ ^FX₄ ^T], [BX₃ ^TX₄ ^F], [BX₃ ^FX₄ ^FDissolve 10 pmol of each in 20 μL of ligation buffer solution and place them on MTP35. These oligonucleotide solutions and buffer solutions were kept at room temperature during the operation of the molecular computer. The ligase and the polymerase for PCR used in the function detect were kept on ice outside the apparatus, and the buffer solution for the PCR reaction was kept at room temperature.
[0164]
The operation of the apparatus in the reaction corresponding to each function will be described step by step.
[0165]
(A) Function "amplify"
The function "amplify" is a function that defines a reaction operation of performing PCR amplification, maintaining the concentration of the original solution, and dividing the solution into a plurality of solutions using the solution of the leftmost argument as a template, as a calculation reaction of the code molecule. Strictly, amplification can only be performed after measuring the oligonucleotide concentration of the original solution.₂Since each oligonucleotide is dissolved in 20 μL of a solution at 5 pmol, each oligonucleotide is dissolved by several pmol in 20 μL of the solution. That is, assuming that 2 pmol per oligonucleotide per solution, the composition of the PCR reaction solution is as follows.
[0166]
Polymerase enzyme (Takara Shuzo) 0.5μL (2.5U)
Amplified DNA solution Contains about 1 fmol of each oligonucleotide
8 μL of dNTP mixed solution (2.5 mM, accessory)
Reaction buffer 10μL (used at 10-fold dilution, accessory)
Primer Prepare 5 pmol of each oligonucleotide on both forward and reverse side
Sterile distilled water was added so that the total volume would be 100 μL
Pyrovest of Takara Shuzo for amplification^TMA DNA Polymerase PCR amplification kit was used. The reaction temperature conditions are as follows.
[0167]
That is,
1. 95 ° C for 30 seconds
2. 50 ° C for 30 seconds
3. 72 ° C 60 seconds 1 to 30 30 cycles
It is.
[0168]
The amount of the PCR reaction solution was changed according to the number of solutions to be divided. After PCR, the PCR product is captured from the reaction solution with streptavidin magnetic beads (Roche Diagnostics). Take 50 μL (0.5 mg) of the magnetic bead stock solution, extract only the magnetic beads therefrom with a magnet, and replace the solution with 50 μL of the B & W solution. The PCR product is captured by mixing the magnetic bead solution and 50 μL of the PCR reaction solution. After the capture, the solution was replaced with 50 μL of the B & W solution, and the temperature was raised to 88 ° C. to dissociate and extract the single-stranded oligonucleotide.
[0169]
(B) Function get
get (T, + S) and get (T, -S) are simultaneously performed by a series of operations.
[0170]
(1) 50 μL of the solution T to be extracted was prepared and discharged to the thermal cycler 38.
[0171]
(2) The first hybridization reaction was carried out by further adding 50 μL of a B & W solution containing 20 pmol of an oligonucleotide in which the complementary strand of the sequence to be captured by T was biotinylated (the following reaction temperature conditions 1, 2).
[0172]
The pipetter takes a liquid from each MTP and proceeds the reaction. The reaction temperature conditions are as follows. That is,
1. 95 ° C for 1 minute
2. 10 minutes at 25 ° C (The temperature is reduced at 10 ° C / minute from 1 to 2)
3. 56 ° C 3 minutes (The temperature is raised at 10 ° C / minute from 2 to 3)
4. 75 ° C for 3 minutes (The temperature is raised at 10 ° C / minute from 3 to 4)
Thus, the thermal cycler was controlled while ensuring the pipetting operation time.
[0173]
(3) In the middle of the temperature of 2, 50 μL of a B & W solution in which 50 μL of the magnetic bead stock solution was dispersed was mixed, and the biotinylated oligonucleotide hybrid was captured on the magnetic beads. The magnetic beads were again held in the thermal cycler 96-well MTP38.
[0174]
The magnetic bead liquid is once sucked into the pipettor of the apparatus, and then retained in a cavity in the pipette tip. At this time, the beads holding the liquid are brought close to a movable permanent magnet attached to a pipettor to collect the beads. During this time, if the solution is drained from the pipette and a new solution is aspirated, the solution can be replaced or the complementary strand of the double-stranded nucleic acid can be separated. In order to disperse the magnetic beads in the solution, pipetting is performed several times with the permanent magnet separated, and the magnetic beads are sufficiently stirred and dispersed in the solution.
[0175]
(4) Next, proceed to the temperature condition of 3. Here, a cold wash step of collecting oligonucleotides that have not hybridized is performed. The oligonucleotide was extracted into 50 μL of the B & W solution at 56 ° C. and output to MTP35. This is the output oligonucleotide solution of get (T, -S).
[0176]
(5) In the hot wash step under temperature condition 4, the temperature was further raised as in 4, and the oligonucleotides captured by the magnetic beads were extracted into a 50 μL B & W solution as in the case of cold wash and output to MTP35. This is the output oligonucleotide solution of get (T, + s).
[0177]
(C) Function merge
Performed with a very simple pipetting operation. This was realized by aspirating each solution to be mixed with a pipettor and collecting and mixing all of them in one MTP well.
[0178]
(D) Function append
This is the reaction shown in FIG. Different oligonucleotides to be applied were reacted simultaneously in different reaction tubes and performed simultaneously.
[0179]
(1) 20 μL of a solution to be subjected to an append reaction is sucked from the MTP 35 with a pipettor and discharged to the thermal cycler 38. In addition, the following solution required for the append reaction was transported to the reaction well 38.
[0180]
For the append reaction, Taq ligase from New England Bio Labs and its dedicated buffer solution were used.
[0181]
The reaction solution is as follows.
[0182]
Taq ligase (NEB) 0.5 μL (20 U)
Reaction DNA solution stock solution 20μl
Reaction buffer 12μL (used at 10-fold dilution, accessory)
Linked oligonucleotides 10 pmol each
Append oligonucleotide 10pmol each
Sterile distilled water was added so that the total liquid volume became 120 μL.
[0183]
(2) A ligation reaction was performed. The control of the thermal cycler was performed as follows. The ligation reaction was performed under the temperature condition 2. The reaction temperature conditions are as follows. That is,
1. 95 ° C for 1 minute
2. 58 ° C 15 minutes (The temperature is reduced at 10 ° C / minute from 1 to 2)
3. 10 minutes at 25 ° C (from 2 to 3 decrease the temperature at 10 ° C / minute)
4. 70 ° C for 3 minutes (increase the temperature at 10 ° C / min from 3 to 4)
5. 74 ° C 3 minutes (Raise the temperature at 10 ° C / min from 4 to 5)
6. 88 ° C 3 minutes (Raise the temperature at 10 ° C / min from 5 to 6)
And
[0184]
(3) After the temperature was lowered to 25 ° C., which is the reaction temperature condition 3, capture by magnetic beads was performed. At this time, the solution in which the magnetic beads were dispersed was discarded, and the ligation buffer solution was directly aspirated from the thermal cycler 38 with a pipettor while the beads were held in the chip. That is, the capture reaction was performed in a ligation buffer solution. Since the buffer solution is not a liquid for capturing, suction and discharge were performed 60 times so that a sufficient amount could be captured. The magnetic beads were held in a thermal cycler 38.
[0185]
(4) Subsequently, the first cold wash was performed. 70 ° C. is a suitable temperature based on the length of the nucleic acid hybrid and the salt concentration of the solution. After the temperature was raised to 70 ° C., the magnetic bead solution was sucked from the thermal cycler 38, the beads were collected with a permanent magnet, and only the solution was discharged to the MTP 35. This solution is a waste liquid.
[0186]
(5) 50 μL of the B & W solution of MTP37 was aspirated with a pipetter, the permanent magnet was released, and the magnetic beads were dispersed and returned to the thermal cycler 38. In the thermal cycler 38, the reaction temperature was set to 5, and a second cold wash was performed. Here, similarly to (4), when the temperature reached an appropriate temperature, suction was performed from the thermal cycler 38, beads were collected with a permanent magnet, and only the B & W solution was discharged to the MTP 35. This solution is also waste liquid.
[0187]
(6) Thereafter, 50 μL of the B & W solution of MTP37 was again sucked with a pipetter, the permanent magnet was released, and the magnetic beads were dispersed and returned to the thermal cycler 38. When the reaction temperature reaches 6 in the thermal cycler 38, the single-stranded oligonucleotide that has undergone the append reaction is dissociated into the solution. This is sucked from the thermal cycler 38, the beads are collected with a permanent magnet, and only the solution is removed. The liquid was discharged onto the MTP 25 and held.
[0188]
(E) Function detect
The reaction shown in FIG. 18 and detection by capillary gel electrophoresis were performed. Graded PCR was performed using a solution that would contain the oligonucleotide representing the finally obtained solution as a template. PCR was performed by dividing the reaction solution for each primer set, and the sequence of the solution was examined by detecting the presence and length of each PCR product.
[0189]
(1) An oligonucleotide solution considered to contain a solution is used as a template. Since the concentration of the oligonucleotide is maintained by the “amplify” in the program, the concentration of the template is estimated therefrom to determine the liquid volume.
[0190]
The composition of the PCR reaction solution is
Polymerase enzyme (Takara Shuzo) 0.5 μL (2.5 U)
DNA solution to be amplified Amount sufficient to contain about 1 fmol of each oligonucleotide
8 μL of dNTP mixed solution (2.5 mM, accessory)
Reaction buffer 10μL (used at 10-fold dilution, accessory)
Primer Prepare 5 pmol of each oligonucleotide on both forward and reverse side
Sterile distilled water was added so that the total volume would be 100 μL
And
[0191]
For the amplification, a Pyrobest DNA Polymerase PCR amplification kit from Takara Shuzo was used. The reaction temperature conditions are
1. 95 ° C for 30 seconds
2. 50 ° C for 30 seconds
3. 72 ° C. 60 seconds 1 to 30 were 30 cycles.
[0192]
The following 12 sets of primers were used in this experiment.
[0193]
(X₁ ^T, [X₂ ^T]), (X₁ ^T, [X₂ ^F]), (X₁ ^T, [X₃ ^T]),
(X₁ ^T, [X₃ ^F]), (X₁ ^T, [X₄ ^T]), (X₁ ^T, [X₄ ^F]),
(X₁ ^F, [X₂ ^T]), (X₁ ^F, [X₂ ^F]), (X₁ ^F, [X₃ ^T]),
(X₁ ^F, [X₃ ^F]), (X₁ ^F, [X₄ ^T]), (X₁ ^F, [X₄ ^F]).
[0194]
These are kept in different wells of MTP35. At the time of the PCR reaction, the primer solution is sucked with a pipettor and discharged to different wells of the thermal cycler 38, and the reaction is completed by adding a solution necessary for the reaction and operating the thermal cycler 38 as set. This can be easily automated. The subsequent introduction of the sample into the capillary by the capillary gel electrophoresis apparatus can also be automated. For capillary gel electrophoresis, a ds1000 gel kit from Peckman Coulter was used. Primer X₁ ^TSince the sample was labeled with FITC, an electrophoresis image was observed. In the present embodiment, the above-mentioned detecting operation was performed manually without performing capillary gel electrophoresis automatically. FIG. 21 shows the result obtained by executing each function of the program by the above method. For example, as a method for identifying a sequence corresponding to the function detect, there is a method as shown in FIG.
[0195]
In order to demonstrate the effectiveness of the computational paradigm of genomic information analysis by a molecular computer, an experiment on DNA computation was performed to actually analyze gene expression information. The calculations can be performed using the basic get, append, amplify, merge, and detect instructions used to solve the 3SAT problem in dynamic programming. The first calculation reaction is an encoding reaction. In a computational reaction that can be performed in a single test tube, information on the transcript of the gene is converted into DCN by an append instruction. The conversion table is represented by an adapter molecule Ai, and is converted 1: 1 to a 2-digit n-ary DCN. The 200 kinds of DCNs can encode up to 10,000 kinds of genes by using them as two-digit 100-digit DCNs. After that, amplification and dispensing reaction to n test tubes are performed by an "amplify" instruction. Finally, a calculation reaction for decoding the DCN is performed on the n test tubes in accordance with the append and get commands. The decoding reaction can be performed in parallel for n tests. When an experiment was performed using DNA oligonucleotides having sequences specific to the transcripts related to the transplanted fragment versus the host disease as input data, it was confirmed that the calculation reaction proceeded specifically and quantitatively.
[0196]
The method of analyzing gene expression information by DNA calculation for DCN has several advantages over the method of analyzing transcript molecules directly on a DNA chip. Since the DCN is converted into DCN having a uniform property and then amplified, it is possible to amplify and analyze without breaking the original frequency distribution. In addition, the same DNA chip for performing the “get” instruction in parallel by DCN decoding can be used as long as it has the same DCN system. In addition, the number of DNA probes is significantly smaller. The labor and cost for producing a DNA chip are greatly reduced. Furthermore, the hybridization reaction of the orthonormalized probe is optimized, and accurate calculation processing can be performed. Also, there is no need to label the transcript molecules, and no errors occur due to variations in efficiency. In addition to these advantages, it is possible to perform not only measurement of an expression profile but also various information analysis on the profile by DNA calculation for DCN.
[0197]
In the above, five basic functions (a function “amplify”, a function “get”, a function “merge”, a function “append”, and a function “detect”) have been described as examples. For example, a function “encode”, a function “decode”, a function “clear”, and the like are also effective in defining calculations in a molecular computer.
[0198]
The function "encode" is a sequence s specific to a test tube T containing molecules._iWhen there is a molecule containing_i, For example, encode (T, s₁, S₂, ..., s_n, C₁, C₂, ..., c_n).
[0199]
The function “decode” is a function for outputting a molecule containing a complementary sequence of the sequence ci adjacent to si when a molecule containing the specific sequence si exists in the test tube T containing the molecule. For example, decode (T, si).
[0200]
The function “clear” is a function for cutting a molecule containing a specific sequence s in the test tube T containing the molecule, and is expressed by, for example, “clear (T, s)”. There are two types of this function "clear". The first function is to cut a molecule in the test tube T at a specific sequence s, and the second is to cut a sequence separated by a specific number of bases from the specific sequence in the test tube T containing the molecule. Function. The function “clear” is defined by the molecular computer 22 as cleavage of a restriction enzyme.
[0201]
Of course, it is not necessary to represent only the eight basic functions including these three basic functions, and other basic functions can be set.
[0202]
Table 2 below shows basic functions, expressions as a processing system in the electronic computer 21, and expressions as a processing system in the molecular computer 22. Note that the processing system in the molecular operation unit 22 is not limited to the one shown in Table 2, but is merely one mode of expressing the processing system in the electronic operation unit 21.
[0203]
As shown in Table 2, basic functions (procedures or functions) are defined in association with the operation of the coding molecule, and variables and constants are defined in association with the coding molecule obtained by encoding the molecule based on the nucleic acid sequence. ing.
[0204]
[Table 2]

[0205]
According to the aspect of the present invention, by utilizing the high parallel computing property of the molecular computer as described above, and by supplementing the function that is difficult to be realized by the molecular computer alone with the electronic computer, the operator can operate the molecular computer. It is not necessary to perform a reaction operation procedure for calculation, assigning of encoded molecules, and the like.
[0206]
In addition, when gene analysis is performed using the present calculation device, it is possible to evaluate the presence or absence of a nucleic acid having a specific sequence as a target, and thereby, for example, the genotype or the state of gene expression Can be determined easily with low experimental error, low cost, and low cost.
[0207]
Furthermore, the present invention also provides the following method and molecular calculation software based on the above description. That is, the present invention provides a molecular computing method characterized by causing an electronic computing unit and a molecular computing unit to function integrally based on molecular information recognizable by an electric program.
[0208]
In addition, the software is applied to a molecular operation device including an electronic operation unit and a molecular operation unit, and is applied to the electronic operation unit and / or the molecular operation unit, and performs a calculation operation by the electronic operation unit and A software for molecular calculation characterized by making the calculation work by the molecular calculation section function in an information format that can be electrically recognized by each calculation section. More specifically, information calculated by the molecular calculation section is converted into an electronic calculation section. Software for molecular calculation characterized by having the function of converting it into an information format that is compatible with the electronic programs of the above. More specifically, the information calculated by the electronic calculation unit is adapted to the calculation work of the molecular calculation unit. An object of the present invention is to provide molecular calculation software having a function of converting information into such an information format.
[0209]
By using the molecular calculation software of the present invention, the above-described computer of the present invention can be easily executed. In addition, the molecular calculation software may manage the computer of the present invention as a whole, independently manage some of its components, or work in conjunction with some of the components in combination. May be managed.
[0210]
II. Genetic analysis application examples
1. Overview
The present inventors have conceived the original idea of treating gene analysis as a calculation using nucleic acid molecules as input data. This idea can be realized using the molecular computer shown by I. For example, converting expressed mRNAs and genes on the genome into encoded nucleic acid data and calculating the logical sum, logical product, and negation of those data to determine genotyping and gene expression conditions in specific diseases Can be.
[0211]
2. Shown below. ~ 4. The first to third examples are realized by applying the molecular computer shown in I.
[0212]
2. First example of applying molecular computer to gene analysis
Examples of problems suitable for the molecular computation device according to one embodiment of the present invention include pure mathematical problems such as the NP complete problem and 3SAT, problems in which input data is given by nucleic acid molecules such as genomic information analysis, and functionality. It is difficult to design molecules and evaluate their functions using an electronic computer.
[0213]
Hereinafter, examples of further genomic information analysis performed by the molecular computer will be described. First, genomic information is converted into a system that is a number represented by an orthonormalized DNA base sequence. Then, the DNA calculation for the DCN is performed as when purely solving a mathematical problem, and the genomic information is analyzed from the calculation result.
[0214]
In this method, two types of nucleic acid probes as shown below, namely, probe A and probe B are used (FIG. 22 (a)).
[0215]
The probe A is composed of a sequence F 'complementary to the base sequence F in a partial region of the target nucleic acid, and a binding molecule bound thereto.
[0216]
Here, the binding molecule is any one of two substances having high affinity specifically to each other. For example, biotin or avidin or streptavidin and the like. Further, the binding molecule may be directly bound to the sequence F ', or may be bound to the sequence F' indirectly through an arbitrary sequence. The arbitrary sequence in the case of indirect binding may be any base sequence or any number of bases. Preferably, it is a sequence that is non-complementary to the base sequence on the target nucleic acid.
[0217]
The probe B is composed of a flag and a sequence S 'complementary to the base sequence S in a partial region of the target nucleic acid. The flag in this example consists of a double strand. The double strand has an arbitrary sequence composed of a plurality of units. Also, the flags do not themselves bind to the target nucleic acid and they need not show any interaction with each other.
[0218]
The sequence F 'and the sequence S' used in the present method have one or more bases, more preferably 15 or more bases.
[0219]
An example of a unit design is shown in FIG. Each unit of the plurality of units of the flag FL may have 10 bases or more, and more preferably about 15 bases. The number of units of the flag FL may be any, but four units are preferable from the viewpoint of easy analysis. However, it is not limited to this.
[0220]
When simultaneously detecting a plurality of target nucleic acids, a flag FL is constructed by combining various types of units. For example, in the case of designing a flag FL consisting of four units of SD, D0, D1, and ED, first, 22 types of units are designed, and two types are selected from them, and an SD unit serving as a primer is selected. , An ED unit as another primer. When D0 and D1 are designed by changing the types of two units to be selected for each type of target nucleic acid using the remaining 20 types of units, it is possible to detect 100 types of different nucleic acid sequences. (FIG. 23 (a)).
[0221]
It is preferable to design the 22 types of units based on orthonormalized base sequences. The orthonormalized base sequences have the same Tm value, and do not form stable hybrids with non-complementary sequences. In addition, a stable secondary structure that does not inhibit the formation of a hybrid with the complementary sequence is not formed. This makes it possible to reduce mishybridization at the time of final detection and to increase the formation speed of hybrids. Therefore, it is possible to improve the detection accuracy and shorten the detection time. In addition, by increasing the number of units or types of units, it is possible to detect 10,000 different nucleic acid sequences.
[0222]
The method of this example will be further described with reference to FIGS. FIG. 22A shows an example of the flag FL composed of four units. The four units are an SD unit serving as a primer in a polymerase chain reaction (hereinafter, referred to as a PCR amplification or a PCR reaction), and a recognition unit for recognizing the type of a target nucleic acid, that is, a D0 unit and a D1 unit. And an ED unit, which is another primer sequence. Each of these units becomes a reading frame in a later step.
[0223]
The detection is performed by first mixing the probe A and the probe B with a target nucleic acid (FIG. 22A). At this time, the target nucleic acid contained in the sample may be a plurality of different nucleic acid molecule groups. For example, if the number of target nucleic acids to be detected is 100 or less, the D0 unit is selected from 10 types of D0-1 to D0-10, and the D1 unit is D1-1 to D1- It is selected from ten types (FIG. 23A).
[0224]
Next, hybridization is performed by incubating the probe A, the probe B, and the target nucleic acid for a certain period of time under conditions suitable for hybridization (FIG. 22 (b)). Hybridization conditions may be as shown in Example 1.
[0225]
By such hybridization, both probe A and probe B bind on the same target nucleic acid (FIG. 22 (b)).
[0226]
Next, the probe A and the probe B hybridized to the target nucleic acid are ligated (FIG. 22C). The conditions for the connection may be as shown in Example 1.
[0227]
Further, it is preferable that the Tm value of the flag FL is designed to be higher than those of the arrays F 'and S'. This makes it possible to prevent denaturation of the flag, which causes a decrease in detection sensitivity, during heating or cooling in operations such as hybridization, ligation, and denaturation in the present detection method.
[0228]
Next, B / F separation is performed on the information of the obtained flag FL (FIG. 22D). Specifically, the binding molecule provided in the probe (A + B) is captured on the solid phase carrier via the binding molecule to be paired with the binding molecule (FIG. 22 (e)).
[0229]
As the solid phase carrier, a substrate, particles such as beads, a container, a fiber, a tube, a filter, an affinity column, an electrode and the like can be used, but beads are preferable.
[0230]
Next, while being captured by the binding molecule, the flag FL of the probe (A + B) is denatured into a single strand (FIG. 22 (f)). PCR amplification is performed on the single-stranded sequence FL 'in the obtained liquid phase (FIG. 22 (g)). As described above, two primer sequences SD and ED are arranged in advance in the flag FL. Therefore, a PCR reaction can be easily performed using this primer sequence. At this time, it is preferable to bind a binding molecule such as biotin to one of the two primers used for PCR, for example, the SD sequence. Detailed conditions of the PCR at this time depend on the designed flag FL.
[0231]
Subsequently, after the completion of the PCR reaction, a double-stranded sequence as a PCR product is recovered by binding the binding molecule to a solid-phase carrier on which the binding molecule is immobilized (FIG. 22 (h)). Here, the solid-phased carrier is the other substance of the binding pair that forms a pair with the binding molecule. Furthermore, the sequence FL 'is removed by denaturation, and only the single-stranded sequence FL is recovered on the solid support (FIG. 22 (i)).
[0232]
Subsequently, the single-stranded flag sequence FL on the solid phase is analyzed. First, the solid support to which the single-stranded flag sequence FL is bound is divided into 10 equal parts (when the D1 unit is D1-1 to D1-D10). To each, one of the D1-1 'to D1-10' sequences and all the D0 'sequences (D0-1' to D0-10 ') bound to the labeling molecule are added and hybridized to the flag sequence FL.
[0233]
Subsequently, the two hybridized nucleic acid molecules are linked by ligation. Here, the ligation conditions and the definition of the labeling substance are as described above. Thereafter, the molecules linked by denaturation are recovered in a liquid phase.
[0234]
Analysis of the obtained labeled nucleic acid molecule can be performed by hybridizing to a DNA chip or DNA capillary on which nucleic acid molecules D0-1 to D0-10 have been previously immobilized. In particular, DNA capillaries that can be processed simultaneously into 10 equal parts, D0-1 to D0-10, will facilitate analysis.
[0235]
For example, when a flag FL is designed using an array of ten types of D0-1 to D0-10 and D1-0 to D1-10, the position of 1 in FIG. 23A corresponds to D0-1. The sequence is fixed and hybridized to the diffusion molecule 63 linked to the labeled D1-1 'molecule. Similarly, at other positions, the corresponding D0 sequence is fixed by a column and hybridized to the diffusing molecule linked to the corresponding D1 'molecule by a row. When such a matrix arrangement is used for a DNA capillary described later (FIG. 23B), analysis can be easily performed.
[0236]
Although an example using ten types of units has been described here, the types of units are not limited to ten types, and may be smaller or larger.
[0237]
As used herein, a "DNA capillary" is a device for detecting a target nucleic acid, in which a complementary sequence to the target nucleic acid is bound, and by binding the target nucleic acid to the complementary sequence. And an apparatus for detecting the target molecule. As shown in FIG. 23 (b), it is possible to simultaneously detect many target nucleic acids by using many DNA capillaries at the same time and arranging different probes in the shaded portions.
[0238]
Further, in the present method, since the orthonormal array is used for each unit of the flag array FL, it is possible to equalize conditions such as the reaction temperature of the hybridization to be performed. Thereby, mishybridization can be prevented, and high accuracy can be obtained. Further, since it is possible to perform many analyzes at once under the same conditions, it is possible to shorten the detection time. In addition, the method makes it possible to convert complex genomic information into a numerical value represented by a DNA base sequence. By performing calculations using DNA molecule reactions, various types of information and complex genes linked to each other can be obtained. Information can be easily analyzed. In addition, since the encoded nucleic acid can be easily amplified after encoding, even a target sequence having a small copy number can be detected accurately and quantitatively. In addition, a large amount of information can be compressed by coding. Therefore, it is possible to save the required number of detection means such as a DNA chip or a capillary array.
[0239]
As used herein, the term "encoding reaction" refers to converting a certain base sequence into a code represented by an orthonormalized base sequence. The above-described steps of FIGS. 22A to 22F correspond to this.
[0240]
Further, the "decoding reaction" used herein means that the code converted as described above is read, thereby restoring the original information. The above-described steps of FIGS. 22A to 22I correspond to this.
[0241]
In this method, it is possible to detect not only one kind of target nucleic acid as described above, but also to detect a plurality of kinds of target nucleic acids at the same time through a similar process by designing a plurality of kinds of flag sequences. It is.
[0242]
3. Second example of applying molecular computer to gene analysis
According to a further aspect of the present invention, there is proposed a general calculation methodology for performing genomic information analysis using the above-described molecular computer. In particular, the genome information analysis method has the following advantages. That is, in such a method, first, an arbitrarily designed desired sequence is arbitrarily assigned to a base sequence of a specific gene. Then, in accordance with the assignment, the base sequence of the specific gene is converted into a designed sequence, and the highly stable sequence obtained after the conversion can be used for the calculation. Therefore, the degree of freedom in designing the reaction is increased, and an accurate reaction can be performed.
[0243]
(A) First embodiment
(1) Outline
A first embodiment of the present invention applied to gene analysis will be described. The first embodiment shows an example of gene analysis for determining the presence or absence of a gene by an operation using a nucleic acid molecule.
[0244]
The outline is as follows. First, a group of cDNAs is prepared based on a group of genes expressed in cells. The information on the expressed genes contained in the obtained cDNA group and the non-expressed genes not contained, that is, the information on the presence or absence of the target gene is converted into the form of a DNA molecule having an artificially designed sequence, and the expression style is changed. change. The DNA molecule obtained by this conversion is hybridized to the nucleic acid for operation. The above process is the process of arithmetic analysis. Here, the DNA molecule functions as a kind of signal that carries information on whether or not a specific target gene is present.
[0245]
For example, by confirming that a certain target gene is present, it can be determined that the gene is an expressed gene. Alternatively, by confirming that the target gene does not exist, it can be determined that the gene is a non-expressed gene. Therefore, in the present analysis method, not only can the target molecule contained in the sample be detected, but at the same time, information that the target molecule not contained in the sample is absent can be obtained. It is.
[0246]
(1.1) Preparation
The calculation method according to one embodiment of the present invention requires the following molecules. Prior to substantial calculations, it is necessary to prepare the following molecules: The preparation can be performed by a method known per se.
[0247]
To detect the cDNA contained in the solution, the two probes shown in FIG._iAnd A_iPrepare a_iIs an oligonucleotide containing a sequence complementary to a part of the sequence of the target cDNA and labeled with biotin at the 5 'end. A_iAre oligonucleotides partially double-stranded by hybridization. A_iIs composed of an artificially designed SD, DCN_iAnd ED at the 3 'end, which is complementary to a partial sequence of the target cDNA and_iIt has a sequence at the 5 'end adjacent to a sequence complementary to the target of the molecule. In addition, the artificially designed base sequence is located at the 3 'end side of the complementary sequence. Also, A_iThe 5 'end of the sequence complementary to the target cDNA is phosphorylated. Double strand A_iThe other strands are SD, DCN_iAnd an oligonucleotide having a sequence complementary to the sequence of ED. a_iAnd A_iIs arbitrarily designed for each target gene to be detected. The “target gene” here is a gene whose presence or absence in a solution is to be detected. At this time, DCN_iAre designed to be different for each target, and SD and ED are all A_iIs designed to be a common array. Since these artificial arrays can be arbitrarily designed, a desired Tm value can be set. Therefore, it is possible to carry out a reaction stably and with less mishybridization.
[0248]
Further, a primer 1 having the same sequence as the SD sequence labeled with biotin at the 5 'end as shown in FIG. 28 and a primer 2 having a sequence complementary to the ED sequence are required (FIG. 28).
[0249]
Further, a DCN indicating “the presence of the target” as shown in FIG._iOligonucleotide 3 (FIG. 31) having a sequence complementary to_i ^*Absent oligonucleotide 6 (FIG. 35) having a sequence complementary to is required.
[0250]
Further, an inverted oligonucleotide 4 as shown in FIG. 32 is required. This is the DCN_iArtificially designed to correspond to DCN_iDCN having a sequence different from that of DCN_i ^*At the 5 'end, and a DCN at the 3' end._iFig. 32 shows an oligonucleotide having a sequence (Fig. 32).
[0251]
(1.2) Conversion to existing and non-existent molecules
In the method of the present invention, first, information that a specific target molecule is present in a sample is converted into “existing molecule”, and information that the specific target molecule is not present in a certain system is “absent”. Into molecules. As used herein, the terms “present molecule” and “present oligonucleotide” are used interchangeably. “Absent molecule” and “absent oligonucleotide” are also used interchangeably.
[0252]
A method of converting the presence or absence of such a molecule into a molecular form expression will be described with reference to FIGS. FIG. 24 to FIG. 45 schematically show molecules present in the system in each step.
[0253]
In the figure, DNA is indicated by an arrow, and the base of the arrow is the 5 'end of the DNA and the tip is the 3' end. A short vertical line to the arrow in the middle of the arrow is a portion indicating a break in the base sequence. In addition, alphabets such as “a”, “A”, and “DCN” shown near the arrows in the figure indicate the names of the sequences. The subscripts “i” and “k” added to alphabets such as “a”, “A” and “DCN” are integers, and are used to indicate which gene each sequence corresponds to. It is attached. Here, an arbitrary sequence is indicated by “i” and “k”. Here, for convenience, “i” indicates an expressed gene and “k” indicates a non-expressed gene. In the figure, when a line is drawn above the sequence name, it indicates a complementary sequence. Shaded circles in the figure indicate biotin molecules, and large white circles represent magnetic beads. The black cross extending rightward from the magnetic beads schematically shows streptavidin molecules that specifically bind to the biotin molecules immobilized on the magnetic beads.
[0254]
a. Conversion of information that "target exists" into "existing molecule"
The conversion of the target to the existing molecule when the target is present is performed by sequentially performing the steps shown in FIGS.
[0255]
First, please refer to FIG. A synthesized as described above_iAnd A_iIs reacted with cDNA in a reaction buffer of an enzyme having high activity at high temperature such as Taq ligase. However, the temperature of this ligation reaction is A_iThe temperature is such that the double-stranded portion of the oligonucleotide does not dissociate. As a result of this reaction, when the target was present, as shown in FIG._iAnd A_iAre concatenated. Next, as shown in FIG. 26, the ligated oligonucleotide is extracted from the reaction solution with magnetic beads having streptavidin bound to the surface. At this time, unreacted a_iMolecules are also captured by the beads but are not involved in subsequent reactions.
[0256]
Subsequently, by applying heat, A captured by magnetic beads_iAnd a_iFrom the connecting molecule of A_iThe complementary strand of the part is separated and extracted (FIG. 27). By this operation, if cDNA is present in the initial solution, the corresponding DCN_iAn oligonucleotide containing a sequence complementary to the sequence is extracted (FIG. 27). Using this extracted oligonucleotide as a template, a PCR amplification reaction is performed as shown in FIG. 28 using the same primer as the SD sequence labeled with biotin at the 5 ′ end and a primer complementary to the ED sequence (FIG. 28). DCN for detecting genes found to be present_iThe sequence is amplified.
[0257]
The double-stranded product from this PCR amplification is captured by streptavidin-coupled magnetic beads as shown in FIG. 29 (FIG. 29). The captured double-stranded PCR product is heated to form a single strand while being captured, and the dissociated complementary strand is removed by buffer exchange (FIG. 30). Subsequently, as shown in FIG._iIs hybridized to the PCR product captured on the beads (FIG. 31). After this hybridization, the excess of oligonucleotides present is removed, and then the DCN captured on the beads by reapplying heat._iIs extracted into the buffer (ie, the present oligonucleotide 3). DCN extracted here_iIs a presence molecule indicating that the target gene is present in the original cDNA solution.
[0258]
b. Conversion of information that "target does not exist" to "absence molecule"
According to the above-described process, the target gene that has been present is converted into a presence molecule (ie, a presence oligonucleotide) that indicates that the target gene is present, and then the information that is not present is converted into a molecule that indicates this (ie, the absence oligonucleotide) ). If the target is not present, an absent oligonucleotide indicating that it was not present is extracted. This extraction is performed as follows.
[0259]
Inverted oligonucleotides as shown in FIG. 32 are prepared in advance for DCNs of all genes to be detected. As described above, the inverted oligonucleotide is DCN_iArtificially designed to respond to DCN_iBase sequence DCN having a sequence different from_i ^*At the 5 'end and adjacent to the 3' end_iOligonucleotide with sequence. By utilizing such a hybridization reaction between the inverted oligonucleotide and the existing molecule, it is possible to convert the “absent target” into a detectable “absent molecule”.
[0260]
First, in the step shown in FIG. 32, the DCN corresponding to the expressed gene extracted in the process of FIG._iThe oligonucleotide 33 is hybridized and an extension reaction is carried out by a polymerase (FIG. 32). As a result, the DCN of the expressed gene_iExtends, DCN_i ^*The complementary strand is synthesized up to the portion of the sequence (FIG. 32). On the other hand, as shown in FIG. 33, the target molecule is a non-expressed gene (here, DCN_kDCN)_kSince the oligonucleotide complementary to is not present in the reaction solution, the inverted oligonucleotide 5 remains single-stranded (FIG. 33). The mixture of these double-stranded and single-stranded can be passed through a column containing hydroxyapatite to extract only the single-stranded inverted oligonucleotide 5 (FIG. 34).
[0261]
DCN corresponding to the non-expressed gene thus extracted_kIs captured on magnetic beads bound with streptavidin (FIG. 35). Next, as in the above-described extraction of only the existing oligonucleotide 3, DCN_k ^*And the excess oligonucleotide is removed to remove the excess oligonucleotide and reveal DCN_k ^*Only the non-existing oligonucleotide 6 hybridized to the above can be extracted (FIG. 35).
[0262]
The step for obtaining the absence oligonucleotide 6 can also be performed as follows. That is, DCN_iA fluorescent molecule such as FITC is labeled at the 5 'end of the oligonucleotide, and DCN_iA hybridization reaction is carried out on a DNA microarray containing a probe having a sequence complementary to Read this with a scanner etc. and check which DCN_iIs detected. At this time, at the same time, the non-existent DCN_kI understand. Therefore, from these data, DCN is used for the next operation._k ^*Is prepared. As described above, the conversion of a non-existing nucleic acid to a nucleic acid having the absence corresponding to the nucleic acid is achieved. This enables a logical operation on the nucleic acid for operation.
[0263]
In the step of producing the non-existent oligonucleotide 6, the non-existent oligonucleotide 6 may be amplified using the single-stranded inverted oligonucleotide 5 as a template. Amplification can be performed, for example, through the steps shown in FIGS. Here, FIGS. 40 to 45 show the respective steps and schematically show the molecules present in each system. Details of symbols and the like shown in each drawing are as described above. First, according to the process of FIG. 34, the inverted single-stranded inverted oligonucleotide 5 is obtained (FIG. 34). As shown in FIG. 40, the DCN bound to the sequence complementary to SD at the 3 'end and to the sequence complementary to the ED sequence at the 5' end, as shown in FIG._k ^*Absent oligonucleotide 6 is extracted by hybridizing oligonucleotide 7 having a sequence complementary to Subsequently, FIG. DCN amplified in the same manner as the existing oligonucleotide 3 by the steps shown in 36 to 22_k ^*Can be obtained. That is, in the step shown in FIG. 41, PCR amplification is performed using the primer 1 having an SD sequence biotin-labeled to the oligonucleotide 7 obtained in the step of FIG. 40 and the primer 2 having a sequence complementary to the ED sequence. (FIG. 41). Next, the PCR product is recovered by binding biotin to the streptavidin molecule (FIG. 42). Subsequently, the PCR product is made into a single strand by thermal denaturation (FIG. 43). Then, DCN_k ^*Absent oligonucleotides having a sequence complementary to are hybridized to the PCR product captured on the beads (FIG. 44). After this hybridization, the excess of oligonucleotides present is removed, and then the DCN captured on the beads by reapplying heat._k ^*Is extracted into the buffer (ie, the absence oligonucleotide 6). DCN extracted here_k ^*Is a non-existent molecule indicating that the target gene is not present in the original cDNA solution.
[0264]
(1.3) Calculation process
In the calculation step, the presence molecule and the absence molecule obtained above are subjected to hybridization and complementary strand synthesis with the calculation nucleic acid described below, thereby solving the calculation formula representing the desired condition, Perform parallel computation to find a solution that satisfies the conditions.
[0265]
Equation 4 is used as an example of the calculation. A specific array DCN₁, DCN₂, DCN₃And DCN₄Is used as a target sequence, and the expression 4 is expressed as a logical expression with respect to the condition of presence or absence of the target sequence. Equation 4 is DCN₁, DCN₂, DCN₃And DCN₄Are shown as desired conditions. That is, in this example, finding the solution of Equation 4 means simultaneously evaluating the presence or absence of a plurality of target sequences in a certain sample.
[0266]
(Equation 4)

[0267]
In the equation, “¬” is a symbol representing “negation”, “∧” is a symbol representing “logical product”, and “∨” is a symbol representing “logical sum”.
[0268]
When the condition set for the nucleic acid for operation is satisfied, the value of the expression becomes “1”, that is, “true”. When the condition set for the nucleic acid for calculation is not satisfied, the value of the expression is “0”, that is, “false”. The present invention also achieves a basic exclusive OR. Therefore, all logical operations of Boolean algebra can be realized by using this combination.
[0269]
FIG. 36 shows the sequence structure of the calculation nucleic acid 8. The calculation nucleic acid is a single-stranded oligonucleotide, and is indicated by an arrow in the figure. The direction of the arrow is from the 5 'end to the 3' end. At the 5 'end is a biotin molecule. The computational nucleic acid includes a plurality of units. The base sequence is M from which the marker molecule binds in order from the 5 'end.₁, DCN₁DCN for detecting oligonucleotides obtained in the absence of₁ ^*, DCN₂A stopper sequence S having a sequence such that the extension of the complementary strand by the polymerase is stopped, and a stopper marker M having a second marker.₂, DCN₃, DCN₄ ^*It is. The sequence of the operational nucleic acid substantially corresponds to the sequence of each term of the logical expression. The "negation" of the logical expression becomes the same array. For example, "DCN₄If the existence of the “” array is “denied”, “DCN₄ ^*Is used. Where DCN₄ ^*Is the DCN as described above₄Is an artificial array designed to correspond to Further, the “OR” symbol may be replaced with an S array. The “logical product” does not need to be replaced as a sequence on the nucleic acid for operation. The value of the calculation of the calculation nucleic acid for Formula 1 becomes 1 when each condition in Table 3 is satisfied. In Table 3, "-" indicates that any state is acceptable.
[0270]
[Table 3]

[0271]
Hereinafter, the steps of the nucleic acid reaction for evaluating the logical formula will be described. The calculation step includes the steps of FIGS. 37 to 39, and the calculation reaction is performed in the following procedure. When performing the above-described operation of the logical expression, an operation nucleic acid having the sequence shown in FIG. 36 is prepared. One kind of the nucleic acid for calculation is put in one tube, and a solution containing the present oligonucleotide and the non-existent oligonucleotide obtained in the previous step is put therein, and a hybridization reaction is carried out (FIG. 37). Here, temporarily, DCN₁, DCN₃, DCN₄Exists and DCN₂If is not present, it is DCN that hybridizes to the computational nucleic acid.₃Only (FIG. 37). In addition, after the hybridization reaction, when an extension reaction is performed under conditions where mishybridization does not occur with an enzyme that is active even at a high temperature such as Taq polymerase, M₂The sequence portion becomes double-stranded and the extension stops at the S sequence (FIG. 38).
[0272]
Next, the computational nucleic acid that has completed the reaction is captured by the streptavidin magnetic beads (FIG. 39). Finally, a hybridization reaction of the marker oligonucleotide is performed as a detection reaction (FIG. 39). FIG. 39 shows a calculation nucleic acid immobilized on a carrier, in which a biotin molecule at the 5 'end is bound to a streptavidin molecule on the carrier. Further, a marker oligonucleotide M having a sequence complementary to the marker detection sequence₁, M₂Prepare A fluorescent molecule is attached to the 5 'end of each of these marker oligonucleotides. In this example, the calculation nucleic acid M₁Since the sequence is not double-stranded, the marker can bind to the computational nucleic acid (FIG. 39). Thereafter, if the unbound marker is removed and the beads are observed under fluorescence, the marker oligonucleotide M₁It can be seen that the value of the logical expression obtained as a result of the operation is "1" because of hybridization and emission of fluorescence.
[0273]
In the above description, the S array is used as the stopper, but the S array is not necessarily required. Alternatively, nucleotides with artificial bases may be included so that the S sequence is eliminated and no complementary strand is formed. At this time, the operation nucleic acid may be arranged on both sides of the S sequence. For example, the S sequence may be designed so that cytosine bases are not included in other sequence portions, and may be designed so that cytosine bases are included in the base sequence of the S sequence. In this case, the extension reaction stops on the S sequence unless dGTP is separately added as a monomer during the extension reaction on the nucleic acid for calculation shown in FIG. Alternatively, a continuous base sequence of guanine or cytosine in which the polymerase easily stops the elongation reaction can achieve the purpose as a stopper. Alternatively, PNA complementary to the S sequence may be hybridized to the nucleic acid for operation. In this case, since a hybrid of DNA and PNA is more stable than a hybrid of DNA and DNA, even a polymerase having 5 'exonuclease activity cannot be removed. Therefore, the S array functions as a stopper.
[0274]
Further, the type of fluorescent label of the marker oligonucleotide may be increased. At present, many fluorescent dyes have been developed, and it is possible to label different nucleic acids with different fluorescent dyes. By doing so, it is possible to label many operation nucleic acids at the same time. For example, in this embodiment, M₁Marker oligonucleotide and M₂If the fluorescent molecule is changed between the marker oligonucleotide and the marker oligonucleotide, it is possible to determine which of the conditions enclosed in parentheses in the logical expression has been satisfied upon detection. In addition, if the marker sequence is changed for each calculation nucleic acid and the fluorescent molecule to be labeled on the marker oligonucleotide is changed, a plurality of calculation nucleic acids can be put into one tube to perform the calculation reaction at the same time. Furthermore, since the fluorescence intensity is proportional to the sufficiency of the logical expression represented by the operation nucleic acid, it is possible to know the magnitude thereof.
[0275]
Further, in obtaining the calculation result, the following can be performed. That is, in the step of amplifying the presence oligonucleotide and the absence oligonucleotide in this embodiment, if the PCR reaction is performed a sufficient number of cycles so that the amplification is saturated, “existence” is changed to “1” and “absence” is reduced. Binary logical operation with “0” is possible. On the other hand, if the number of cycles is suppressed so that the PCR reaction does not saturate the amplification and is obtained in an amount proportional to the abundance of the original cDNA, the logical expression can be evaluated stochastically in the interval [0, 1]. it can. For example, in the case of an expressed gene, a result corresponding to the expression level can be obtained, and in the case of a genomic sequence, the difference between heterozygous or homozygous can be known from the operation result.
[0276]
Further, the nucleic acid for operation may be fixed on the substrate in the form of a small spot like a DNA microarray, and the location may be addressed so that the logical expression of the nucleic acid for operation corresponds. In such a case, it is preferable that the marker oligonucleotide be fluorescently labeled. When the above-described operation reaction is performed using the operation nucleic acid on the microarray, the operation result can be read by a DNA microarray reading scanner.
[0277]
The marker oligonucleotide may be labeled with a biotin molecule. At this time, no biotin is attached to the 5 'end of the computational nucleic acid, and the 3' end and the 5 'end contain a restriction enzyme recognition sequence for cloning. Furthermore, in the reaction, the step of capturing the nucleic acid for calculation using streptavidin magnetic beads as shown in FIG. 39 is not performed. In the detection reaction in this case, after the marker oligonucleotide is subjected to the hybridization reaction, both the marker oligonucleotide and the non-operating nucleic acid hybridized with the nucleic acid for calculation by streptavidin magnetic beads are captured, and the captured nucleic acid for calculation is cloned. Then, the base sequence is read with a sequencer. Thereby, the nucleic acid for operation whose operation result is “1” can be confirmed. In this way, a plurality of types of arithmetic nucleic acids can be put into one tube to perform a reaction.
[0278]
Here, an example using four target sequences is used, but more target sequences can be used. Here, biotin and streptavidin were used as tags for recovery, but the present invention is not limited to this, and any substance having a high affinity for the tag and the tag may be used. May be.
[0279]
(B) Second embodiment
According to a preferred embodiment of the present invention, in the method described above, an orthonormal array may be used as the DCN array. "Orthonormalized sequence" is an artificially designed base sequence, and "normal" is the melting temperature (T_m), And "orthogonalization" means that no mishybridization occurs and a stable structure is not formed in the self molecule.
[0280]
For example, to obtain an orthonormal sequence of 15 bases, any 5 bases are randomly generated. These short base sequences are called "tuples". 5 base tuple is 4⁵= 1024 types. Three of these tuples are selected and ligated to form 15 bases. A tuple complementary to the tuple used for this connection is not used for subsequent connections. Here, the T base of the 15 base sequence_mMake a set of 15 bases such that are within ± 3 ° C. Also, it is calculated whether a stable structure is formed in the self-molecule, and such a 15 base is excluded if a stable structure is formed.
[0281]
Finally, it is verified whether all sequences form stable double strands with each other. The sequence of 15 bases generated by the above method is more preferable since it does not form a hybrid with each other if the reaction temperature is appropriately selected, and performs an independent hybridization reaction even when mixed. Nucleic acid a such that these orthonormalized sequences correspond to a specific gene base sequence_iAnd A_iWhen the reaction is performed in accordance with the first embodiment by selecting the above sequence, the reaction conditions become simpler and more accurate calculation reaction can be performed.
[0282]
(C) Third embodiment
Subsequently, according to a preferred embodiment of the present invention, it can also be used to determine a genotype determined by the presence of a specific nucleotide sequence at each of a plurality of loci. Here, an example of a method for determining such a genotype will be described.
[0283]
A logical expression corresponding to the genotype is set, and an operation nucleic acid is designed based on the logical expression. Thereby, the genotype can be determined without using an electronic computer or a determination table. Specifically, for example, it can be determined as genotype A when it is base A at locus 1, base T at locus 2 and base G at locus 3, and base A at locus 1 and base A at locus 2. C, if it is base T at locus 3, it is genotype B, and if it is base A at locus 1, base C at locus 2, and genotype C if it is sequence G at locus 3, , And the logical expressions satisfying each are as shown in the rightmost column of Table 4.
[0284]
[Table 4]

[0285]
In genotyping, an arithmetic reaction is performed using an arithmetic nucleic acid corresponding to each logical expression. First, for each element of the expression, the value is 1 when a specific sequence is present at the locus, and 0 when it is absent, and if there is an expression whose final value is 1 as a whole, the expression The corresponding genotype is the determination result. According to the method of the present invention in which the calculation is performed using such a nucleic acid, typing of a gene having a large number of loci and a very small number of genotypes can be easily performed without requiring a complicated table or a computer.
[0286]
(D) Fourth embodiment
Further, according to a preferred embodiment of the present invention, the above-described method can be used for examining the expression and non-expression conditions (ie, conditions) of various genes in gene expression of cancer cells. . In contrast to the above-described first embodiment, the third embodiment can be called a so-called “inverse problem”.
[0287]
In advance, operational nucleic acids representing various logical expressions are prepared. This arithmetic nucleic acid is composed of various logical formula elements phosphorylated at the 5 'end, that is, DCN_i, DCN_i ^*, S, M, etc. In addition to these, for example, a complementary nucleic acid 9 for connection for connecting logical formula elements as shown in FIG. 45 is prepared. The sequence of the complementary nucleic acid 9 for linking may have a sequence complementary to the sequence of the two portions existing adjacent to the partition portion desired to be linked.
[0288]
If these nucleic acids are hybridized and ligated with a ligase, an operational nucleic acid to which a logical formula element is arbitrarily linked can be obtained. By designing with sufficient consideration of the sequence of the complementary nucleic acid for ligation, it is possible to prevent an inappropriate logical formula from being generated.
[0289]
In order to solve the inverse problem using the nucleic acid for operation, cDNA obtained from cancer cells is converted into an array of logical formula elements according to the first embodiment, and various logical formulas prepared in advance are converted. A logical operation is performed using the indicated operational nucleic acid. Finally, it is detected whether or not any of the logical expressions represented by these operational nucleic acids is satisfied. Of these, by interpreting the meaning of the nucleic acid for operation in which the value of the logical expression is 1, it is possible to clarify what kind of gene expression or non-expression condition is satisfied. . Alternatively, it is possible to at least clarify the partial conditions.
[0290]
In order to identify the sequence of the arbitrarily prepared arithmetic nucleic acid, first, the reaction is performed in one container. Next, the nucleic acid for operation is collected on streptavidin magnetic beads using the marker oligonucleotide of the first embodiment to which the biotin molecule has been bound. Subsequently, it is preferable to perform sequencing and read a logical expression by a method of reading the contents of the operation molecule. Also, instead of arbitrarily preparing a calculation nucleic acid, a method of synthesizing a calculation nucleic acid by determining a logical formula and coupling may be used. In this case, these logical expressions may be fixed by addressing them on the DNA microarray, and the logical expressions may be confirmed at the time of detection. Alternatively, the reaction may be performed by placing one kind of arithmetic nucleic acid in one container.
[0291]
By comparing these results with normal cells, it is possible to easily identify a causative gene of a disease such as a cancer caused by a combination of unknown genomic nucleotide sequences or a genetic abnormality caused by a combination of unknown gene expressions. .
[0292]
(E) Discussion
There is no conventional technology capable of converting the absence of a specific nucleic acid into a nucleic acid expression, and such a concept is not considered. For example, a DNA microarray is often used to detect the expression state of mRNA. In this microarray, oligo DNA designed based on a known gene base sequence or cDNA obtained in advance is fixed in an array on a slide glass as a probe.
[0293]
In order to detect expression with this microarray, a fluorescently labeled cDNA is prepared from mRNA, a hybridization reaction is performed between the probe of the microarray and the cDNA, and a specific labeled cDNA is immobilized at a place where a probe of a specific sequence is immobilized. Detect by combining and shining. However, since a labeled cDNA cannot be prepared from mRNA that is not expressed, it cannot be detected that the gene is not expressed. That is, mRNA may be lost during the experiment, or it may be determined that the gene is absent because the amount of fluorescently labeled cDNA generated is small despite expression of the gene. Unlike such a conventional method, according to one embodiment of the present invention, it is possible to visualize information on an absent nucleic acid.
[0294]
When reacting gene nucleic acids, unexpected nucleic acids may form a hybrid and react when many genes are mixed. The number of bases such as guanine and cytosine contained in the base sequence stabilizes the structure of the double-stranded nucleic acid. If such bases are different for each gene nucleic acid to be handled, the optimal temperature for forming a hybrid differs. Therefore, not all nucleic acids can form an appropriate hybrid without mismatch. In addition, when nucleic acids having a base sequence that has a structure within the self molecule and has low reactivity with the target sequence are mixed, the reaction that is theoretically expected may not proceed. Compared with such a conventional method, in the method according to the embodiment of the present invention, information is replaced with a nucleic acid molecule designed under preferable conditions and then used in the reaction, so that a stable reaction can be performed. is there.
[0295]
Although a detection method using an enzyme such as chemiluminescence has high sensitivity, the process for detection is troublesome and time-consuming. Further, it is difficult to perform a plurality of types of chemiluminescence in one tube. In addition, when a calculation is performed using a nucleic acid for calculation, only one kind of nucleic acid for calculation can be reacted in one tube in a single color development or luminescence reaction to see the result. In contrast, according to the embodiment of the present invention, it is possible to detect a large number of target nucleic acids simultaneously with high sensitivity.
[0296]
Further, in the conventional method, a logical expression created based only on the presence condition of a nucleic acid is rewritten into a calculation nucleic acid. However, the problem that exists in the world is the so-called “reverse problem”. For example, when examining the expression state of each gene in a genetic disease, the problem is what kind of gene nucleic acid will be sick if conditions for the presence or absence of the gene nucleic acid are satisfied. Therefore, the problem is to find a logical expression for the presence and absence of the gene nucleic acid. Conventionally, such a problem has been solved by performing cluster analysis on data of the DNA microarray using a large computer. Therefore, a great deal of time and money was required. According to an embodiment of the present invention, such a problem can be solved in a short time and economically.
[0297]
Here, a method for gene analysis has been described, but the present invention is not limited to this, and information processing by various parallel calculations can be performed without exceeding the scope of the present invention. That is, even if a mathematical problem of information other than gene analysis, for example, a mathematical problem is solved by performing parallel processing, it will be easily understood by those skilled in the art who can obtain excellent advantages.
[0298]
III. Other application examples
As described above, one embodiment of the present invention
A molecular computer that performs molecular calculation by a chemical reaction of the molecule,
A molecular control unit comprising: a reaction control unit that controls a reaction of a chemical reaction of a molecule; and an automatic detection unit that generates a chemical reaction of the molecule based on the reaction control of the reaction control unit and detects a reaction result of the chemical reaction. When,
A variable and a constant defined in association with a coding molecule obtained by encoding a molecule based on a nucleic acid sequence, and a program described by a function defined in association with the operation of the coding molecule, the operation of the coding molecule A control command generating unit that converts the control command into a control command for driving the reaction control unit that performs a reaction control of the chemical reaction of the corresponding molecule and generates a procedure of the control command.
An electronic calculation unit having an output unit for outputting a procedure of the control instruction
Molecular computing device equipped with
I will provide a.
[0299]
In this embodiment, among the chemical reactions of molecules, attention is paid to the regularity of the reaction of the nucleic acid sequence, and the regularity is associated with the arithmetic reaction. Similar molecular calculations can be performed by associating not only the regularity of the reaction of nucleic acid sequences but also chemical reactions that occur based on other biological and biochemical regularities of molecules with arithmetic reactions.
[0300]
For example, biological and biochemical substances other than the nucleic acid sequences of the present invention can be used, for example, biological elements such as peptides, oligopeptides and polypeptides, proteins, and antigens and antibodies. . In this case, these elements are encoded in place of the nucleic acid sequence, and the variables and constants defined in association with each of the encoded elements are defined in association with the operation reaction of the encoded elements. A source program may be created by a program described by a function.
[0301]
This specification confirms that the following inventions are included.
[0302]
A molecular computation plan design device that sets a plan for molecular computation that performs molecular computation by a chemical reaction of the molecule,
A variable and a constant defined in association with a coding molecule obtained by encoding a molecule based on the biological or biochemical properties of the molecule, and a function defined in association with an operation reaction of the coding molecule. A control command generation unit that converts the program into a control command for driving a reaction control unit that performs a reaction control of a chemical reaction of a molecule corresponding to the arithmetic reaction of the code molecule, and generates a procedure of the control command.
An output unit that outputs a procedure of the control instruction;
[0303]
IV. Application example of molecular calculation section
The configuration of the molecular operation unit 22 shown in FIG. 3 is only an example.
[0304]
For example, by applying a detection device suitable for fully automatic processing shown in the following (1) to (10) to the molecular operation unit 22, the processing speed can be improved at the same time as labor saving. The same components as those of the above-described embodiments are denoted by the same reference numerals, and detailed description thereof will be omitted.
[0305]
(1) Detector using FCS
A detection apparatus using FCS (Fluorescence Correlation Spectroscopy) is disclosed in, for example, JP-A-2001-269198 and JP-A-2002-281965.
[0306]
FIG. 46 is a diagram illustrating an example of a detection device using a confocal microscope using FCS. Excitation light is emitted from a laser light source 41, reflected by mirrors 54a and 54b, and passed through a filter 42 (IF). The laser light is condensed by a condenser lens 43, and a sample is collected by a dichroic mirror 44 (DM). One of 45 is irradiated. Thus, the fluorescent substance in the sample 45 is excited by the laser light and emits fluorescent light. This fluorescence emission passes through the pinhole 49 via the filter 46 (F), the lens 47, the mirror 54c, and the lens 48. As a result of passing through the pinhole 49, only the fluorescence emitted from the focused fluorescent substance in the sample 45 is amplified by the photomultiplier tube 50 (PMT). The amplified signal is output to the digital correlator 53 via the amplifier 51 and the discrimination circuit 52. The digital correlator 53 analyzes the received fluorescence signal by FCS.
[0307]
With this detector, the intensity of the fluorescent light emitted by an average of several (one in some cases) fluorescent substances is measured for a certain period of time, the autocorrelation function of the fluctuation of the fluorescence originating from Brownian motion is obtained, and this autocorrelation function is analyzed. Thereby, various data on the fluorescent substance are obtained.
[0308]
For example, the laser light source 41, the photomultiplier tube 50, the discriminator 52, the digital correlator 53, and the like are controlled by the automatic controller 151, and the sample 45 is transported by using the transport mechanism 160 controlled by the automatic controller 151. . The detection result obtained by the digital correlator 53 is sent to the electronic calculation unit 21 via the communication unit 24 as a detection result of the detection unit 158.
[0309]
For example, with a configuration in which the sample 45 is dropped on each well 45a of a multi-well plate having a plurality of wells 45a made of glass on the bottom surface, the reaction container and the detection container are realized using the same well. This makes it easy to automate the detection device with high sensitivity.
[0310]
FIG. 47 is a diagram showing the analysis principle of FCS.
[0311]
As shown in FIG. 47A, a minute measurement target area 55 is irradiated with laser light. The fluorescence intensity of the emitted fluorescent light is measured over time from the fluorescent substance present in the minute area, and data as shown in, for example, FIGS. 47 (b) and 47 (c) is obtained. Then, an expected value of the product of the fluorescence intensities I (t) and I (t + τ) at two different points in time is calculated, and an autocorrelation function G (τ) = <I (t) I (t + τ)> is obtained.
[0312]
Then, the obtained autocorrelation function G (τ) is analyzed, and the autocorrelation function of each nucleic acid probe before and after addition is compared.
[0313]
The concept of such an analysis becomes clearer from FIGS. 47 (b) and 47 (c).
[0314]
As shown in FIG. 47 (b), when the nucleic acid probe is not bound to the target allele, the function of fluorescence intensity I (t) is large because the size of the molecule is small, and the speed of Brownian motion is large. The frequency of (t) is large.
[0315]
On the other hand, when the nucleic acid probe binds to the target allele, data having a small frequency is obtained as shown in FIG. 47 (c). Therefore, by comparing the autocorrelation functions obtained based on the fluorescence intensity, it becomes clear whether or not the probe has bound. In order to determine which nucleic acid probe has been bound by the FCS, for example, each nucleic acid probe may be distinguished by a fluorescent label having a different excitation wavelength or fluorescence wavelength.
[0316]
If the size of each nucleic acid probe is different, the change in Brownian motion and the autocorrelation function are different. Therefore, it is possible to determine which nucleic acid probe has bound by using nucleic acid probes having different sizes.
[0317]
(2) Detection device using DNA capillary array
A detection device using a DNA capillary array is disclosed in, for example, JP-A-11-75812.
[0318]
FIG. 48 is a diagram showing an example of the DNA capillary 70. As shown in FIG. 48, various stripes are formed along the flow path on the inner wall 72 of the capillary 71 having, for example, a glass light-transmitting and cylindrical injection opening 74 and a discharge opening 75 at both ends. Are sequentially immobilized by the photoreaction of the capping agent. In order to perform the measurement, a sample containing single-stranded DNA labeled with fluorescence is introduced through the injection opening 74 of the capillary 71, and reacted with each of the DNA probes 73a, 73b, 73c. . By measuring the position where the capillary 71 shines, the sequence of the DNA contained in the sample can be determined.
[0319]
FIG. 49 shows a capillary array 75 in which DNA capillaries 70 are integrated. FIG. 49A is a perspective view, and FIG. 49B is a top view.
[0320]
A substrate 81 made of a material such as glass or silicon is provided with a groove having a pattern as shown in the figure, whereby a plurality of DNA capillaries 82a, 82b, 82c... Are formed. One end of each of the DNA capillaries 82a, 82b, 82c... Extends right below the individual entrances 83a, 83b, 83c... And is open to the outside air through the individual entrances 83a, 83b, 83c. They are connected together by a connecting flow path 86 as a road, extend to a position directly below the common entrance 84, and are opened to the outside air through the common entrance 84. The individual entrances 83a, 83b, 83c... And the common entrance 84 can be attached by, for example, bonding an adhesive in a thin film shape by a screen printing technique.
[0321]
The DNA probes 85a, 85b, 85c... Are, as shown in the drawing, coaxial straight lines 87 perpendicular to the direction in which the fluid flows along the flow paths of the DNA capillaries 83a, 83b, 83c.₁, 87₂, 87₃, ... arranged on the top. A different sample is simultaneously injected from each of the individual entrances 83a, 83b, 83c... And discharged to the common entrance 84. For example, a sample is injected from a sample holding array in which 96-well sample holding containers are provided in an array shape into individual entrances 83a, 83b, 83c,... By a hose-shaped injection mechanism provided corresponding to each holding container. . Thereby, many types of samples can be processed simultaneously. In addition, since the hybridization reaction can be performed only by injecting the reaction solution into each capillary, speedup and automation can be achieved by parallel processing. For example, the transfer mechanism 160 arranges the hose-shaped injection mechanism at each of the individual entrances 83a, 83b, 83c,... Under the control of the automatic control unit 151. Further, a fluorescence detection device is provided as a detection unit 158 in correspondence with each of the DNA capillaries 83a, 83b, 83c..., And the fluorescence detection device simultaneously scans the DNA capillaries 83a, 83b, 83c. Go. By analyzing the scan position and the fluorescence detection position in association with each other, it is possible to automatically detect which probe of which capillary emits light. The analysis result is transmitted from the detection unit 158 to the electronic calculation unit 21 via the communication unit 24. This allows simultaneous processing of other types of samples including detection.
[0322]
(3) Detection device using microarray in plate
A detection device using the in-plate microarray is shown in, for example, JP-T-2001-510339.
[0323]
FIG. 50 is a diagram showing an example of the configuration of a detection device 90 using a microarray in a plate. A large number of holes are provided in an array to form a reaction vessel 91, and different DNA probes 95 having a known sequence are immobilized on the bottom of each of a multi-hole plate 92 provided with a plurality of the reaction vessels 91. Then, a solution containing a fluorescently labeled single-stranded DNA molecule is injected into each reaction vessel 91 and hybridized. Then, the fluorescent light detecting mechanism 93 detects which position is illuminated. Thus, the sequence of the DNA contained in the solution can be determined.
[0324]
This fluorescence detection mechanism 93 corresponds to the detection unit 158 in FIG. The fluorescence detection mechanism 93 includes a base 93a and a lid 93b. The base 93a is provided with a plate disposing portion for disposing the multi-hole plate 92. Further, a CCD 93c is arranged on the lower surface of the plate disposing portion. When the multi-hole plate 92 is arranged on the CCD 93c, the lid 93b is closed, and light is emitted from the illumination device 93d provided on the lid 93b, the hybridized reaction vessel 91 emits light. This light emission is detected by the CCD 93c and analyzed by the CCD imager 93e in the base 93a. The analysis result is transmitted to the electronic calculator 21 via the communication unit 24 (not shown). The CCD 93c and the illuminating device 93d may be arranged two-dimensionally in an array for each reaction vessel 91, or may be arranged one-dimensionally so as to detect one row in the array, as shown in FIG. The CCD 93 c and the illuminating device 93 d may be scanned on the rail 96 by a mechanism to detect all the reaction vessels 91. For example, the CCD 93c, the illumination device 93d, and the CCD imager 93e are automatically controlled by the automatic control unit 151.
[0325]
When the same multi-well plate as the reaction vessel in the above embodiment, for example, a 96-well plate is used, detection can be performed immediately after the reaction, so that the processing can be sped up.
[0326]
(4) Detection device using ECA (Electronic Chemical Array)
Detection devices using ECA (Electronic Chemical Array) include, for example, JP-A-6-245800, JP-A-5-285000, JP-A-5-199898, JP-A-2000-83647, JP-A-2002-308871, and JP-A-2002-214192. Is shown in
[0327]
FIG. 51 is a diagram illustrating an example of the configuration of the ECA chip 100. The ECA chip 100 is a chip formed of, for example, a ceramic in which a synthetic oligo DNA probe 101 of 80 bases is covalently bonded to a microelectrode 102 made of gold or the like.
[0328]
FIG. 52 is a diagram for explaining a detection principle using an ECA chip. As shown in FIG. 52, on the fixed probe electrode 102 on which the synthetic oligo DNA probe 101 is immobilized, the sample DNA 103 is bonded to the synthetic oligo DNA probe 101 by being brought into contact with the sample solution to generate double-stranded DNA. When the intercalator 104 is incorporated into this double strand, current flows only to the electrode 102 bound to the paired oligo DNA (103), and no current flows to the unbound electrode 102 (104). It is electrically detected by the detection unit 158 connected to each electrode 102 and transmitted to the electronic calculation unit 21 via the communication unit 24. The detection operation by the detection unit 158 is controlled by, for example, the automatic control unit 151.
[0329]
As described above, since a pairing reaction can be detected as an electric signal, a laser scanning device is unnecessary. In addition, since the oligo DNA is stuck on the microelectrode 102, and a correlation is obtained between the amount of the double strand formed and the amount of the intercalator 104, it is possible to quantitatively detect the DNA.
[0330]
As described above, it is possible to perform label-free and high-speed detection by the detection device using the ECA chip. In addition, it is a low-cost product that has high sensitivity, does not require PCR, and can be reused, and can reduce chip cost.
[0331]
(5) Detection device using an electrode-immobilized DNA probe array
Detection devices using an electrode-immobilized DNA probe array are shown in, for example, JP-A-9-504910 and JP-A-9-503377.
[0332]
FIG. 53 is a diagram showing a detection method of a detection device using an electrode-immobilized DNA probe array. First, as shown in FIG. 53 (a), DNA probes 108a, 108b and 108c are immobilized on a plurality of electrodes 107 formed on a substrate 106. At this stage, no potential is applied to the electrode 107 or a negative potential is applied to the electrode 107. Then, as shown in FIG. 53 (b), each electrode 107 is set to a positive potential by the voltage source 109 controlled by the automatic control unit 151 to attract a DNA molecule having a negative (-) charge, thereby enhancing the hybridization effect. After the hybridization reaction, as shown in FIG. 53 (c), each electrode 107 is kept negative (-) by the voltage source 109 to exclude those other than DNA strongly bound to the DNA probes 108a, 108b, 108c. And perform detection. For detection, a fluorescent label or the like is applied to the DNA molecules in the solution in advance, and this is measured by a detection unit 158 such as a fluorescence detection mechanism.
[0333]
As described above, according to the detection device using the electrode-immobilized DNA probe array, the hybridization reaction can be performed at high speed, unlike a simple microarray.
[0334]
(6) Detection device using microbead array
Detection apparatuses using an optically identifiable microbead array are shown in, for example, JP-T-2001-520323, JP-T-2002-534657, and the like.
[0335]
FIG. 54 is a diagram showing a detection principle of a detection device using a microbead array.
[0336]
First, as shown in FIG. 54 (a), a microtube 114 containing a solution in which probe DNA 112 having a known sequence is immobilized on microbeads 111 of a first color that can be optically individually identified is prepared. At this time, the first color of the microbeads 111 is made to correspond to the immobilized probe DNA 112. Then, the sample DNA molecule 113 fluorescently labeled with the second color is mixed with the solution in the microtube 114, and the sample DNA molecule 113 and the probe DNA 112 are hybridized. Next, as shown in FIG. 54 (b), the solution is dropped one by one from the tip of the microtube 114, and the falling solution is irradiated with the first laser light from the first laser irradiation source 115, and this is irradiated with light. The diameter and the color of the microbeads 111 are measured by detecting using the sensor 117, and the second laser light is irradiated from the second laser irradiation source 116 to the dripping solution so as to intersect with the first laser light. The amount of fluorescence on the surface of the microbeads 111 is measured by detecting with 118. The measurement signal is transmitted to the electronic calculator 21 via the detector 158 and the communication unit 24. The presence or absence of the microbeads 111 can be detected by the first laser light to detect what kind of probe DNA 112 is, and the presence or absence of hybridization of the sample DNA molecule 113 can be detected by the second laser light. If the sample DNA molecule 113 is not observed by the second laser light, it indicates that hybridization has not occurred, and thus the sequence of the DNA molecule in the solution can be determined. The automatic control unit 151 controls the first laser irradiation source 115, the optical sensor 117, the second laser irradiation source 116, the photomultiplier tube 118, and the detection unit 158.
[0337]
Further, according to the detection device using the microbead array, the detection can be performed in a flow system, and therefore, the detection can be performed at high speed and with high sensitivity.
[0338]
(7) Detection device using microtransponder
A detection device using individually identifiable microtransponders is shown for example in US 6046003.
[0339]
FIG. 55 is a diagram illustrating an example of the detection device 120 using a microtransponder. As shown in FIG. 55, the metal coil antenna 121 is covered around the flow cell 122 of the flow cytometer. The individually identifiable transponder 123 passes through the flow cell 122 and is decoded by the scanner device 124. A signal carrying data transmitted from the transponder 123 is amplified by an amplifier and processed by a scanner device 124. When the transponder 123 is decoded, the fluorescence from the transponder 123 is detected and analyzed by the flow cytometer using the laser light from the laser irradiation source 120a. The processing results and analysis results of the scanner device 124 and the flow cytometer are transmitted to the electronic calculator 21 via the communication unit 24. The laser irradiation source 120a and the scanner device 124 are controlled by the automatic control unit 151.
[0340]
As described above, according to the detection device 120 using the microtransponder, the detection can be performed in a flow system, and thus the detection can be performed with high speed and high sensitivity.
[0341]
(8) Detection device using frequency conversion element
A detection device using a frequency conversion element is disclosed in, for example, Japanese Patent Laid-Open No. 6-94591.
[0342]
FIG. 56 is a diagram illustrating an example of a detection device 125 using a frequency conversion element. The probe DNA 126a is solid-phased on the electrode 126 provided on the frequency conversion element 127. Then, the electrode 126 is immersed in the sample solution to form a hybrid with the sample DNA 126b in the sample solution. The detector 128 detects a frequency change of the frequency conversion element 127 accompanying a weight change due to the formation of the hybrid, and detects the presence or absence of the hybrid. The detection result of the detector 128 is transmitted to the electronic calculation unit 21 via the communication unit 24. The detector 128 is controlled by the automatic control unit 151.
[0343]
As described above, according to the detection device 125 using the frequency conversion element, label-free detection can be performed with high sensitivity.
[0344]
(9) Detection device using surface plasmon resonance
A detection device using Surface Plasmon Resonance is disclosed in, for example, JP-A-2002-51774.
[0345]
FIG. 57 is a diagram illustrating an example of the detection device 130 using surface plasmon resonance. The probe DNA 132 is fixed on the back surface of the sensor chip 131. Then, the sample DNA 133 is introduced into the back surface of the sensor chip 131 by the flow cell 135, and the probe DNA 132 and the sample DNA 133 are hybridized. In this state, the metal film 131a formed on the surface of the sensor chip 131 on which the probe DNA 132 is not fixed is irradiated from the light source 134 so that polarized light is totally reflected via the prism 134a. Then, the detection unit 158 detects the light reflected on the surface of the sensor chip 131. The detection result of the detection unit 158 is transmitted to the electronic calculation unit 21 via the communication unit 24. The light source 134 and the detection unit 158 are controlled by the automatic control unit 151.
[0346]
As a part of the detected reflected light, a part where the reflected light intensity is reduced is observed. The angle at which the dark portion of the light appears depends on the mass on the surface of the sensor chip 131. When the sample DNA 133 is added to the ligand and a binding reaction occurs, a change in mass occurs, and the dark portion of light shifts from I to II in the figure. If the mass is reduced by dissociation, the shift from II to I is reversed. In the detection device 130, the amount of shift from I to II in the figure, that is, the change in mass on the surface of the sensor chip 131 is taken on the vertical axis, and the time change of mass is acquired as measurement data. Thereby, the interaction between the probe DNA and the sample DNA can be detected in real time with high sensitivity without labeling the probe DNA.
[0347]
In addition, after immobilizing a plurality of different probe DNAs 132 on the same chip 131 and reacting them with the same sample solution, selectively irradiating only the probes to be measured with light sequentially and measuring them, Can be analyzed at the same time.
[0348]
(10) Detector using yarn chip
FIG. 58 is a diagram illustrating an example of a configuration of a detection device 140 using a thread tip. The probe DNA 142 is immobilized on a plurality of positions of the thread member 141. Then, the thread-like member 141 on which the probe DNA 142 is immobilized is wound around the core 143. The core 143 around which the thread-like member 141 is wound is placed in a microtube 144 containing a sample solution 145, hybridized, and washed. The washed microtube 144 is irradiated with light from a light source 146, and the reflected light is detected by a detection unit 158. The detection result is transmitted to the electronic calculation unit 21 via the communication unit 24. The detection result may be analyzed once by a computer, and the analysis result may be transmitted to the electronic calculator 21. The light source 146 and the detection unit 158 are controlled by the automatic control unit 151.
[0349]
In this way, by winding the probe DNA around the thread-like member 141, the density of the probe DNA can be increased. Also, there is an advantage that the detection device can be easily fully automated.
[0350]
Instead of immobilizing DNA probes, sample DNAs from a plurality of different patients may be immobilized. In particular, since the measurement data is aligned along the thread winding trajectory, mistakes hardly occur, and high-speed analysis can be performed for each address.
[0351]
This embodiment is not limited to the detection devices described in (1) to (10). The detection devices shown in (1) to (10) can be used in combination.
[0352]
For example, the DNA capillary shown in (2) and the microbeads shown in (6) can be used in combination. FIG. 59 shows an example of a bead capillary array in which microbeads are arranged in a capillary. (A) is a top view and (b) is a cross-sectional view. Specifically, a plurality of micro beads 111 are arranged in the capillary 147. The presence or absence of sample DNA in the sample can be detected by introducing the sample solution into the capillary 147 in which the microbeads 111 are formed and observing the sample solution with an optical microscope. Thus, there is provided a detection device in which the capillary shown in (2) and the microbeads shown in (6) are combined. Further, other combinations of (1) to (10) can be applied. For example, (1) and (2), (2) and (10), (2) and (6), (3) and (3) 9), (4) and (6), (4) and (10), (7) and (10), (2) and (4) and (6), (2) and (7) and (10) The preferred combination is as follows.
[0353]
【The invention's effect】
As described in detail above, according to the present invention, by utilizing the high parallel computing performance of a molecular computer and complementing functions that are difficult to be realized only by a molecular computer with an electronic computer, the present invention provides faster processing than a conventional electronic computer. Calculations can be performed.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a molecular computer according to the present invention.
FIG. 2 is a schematic diagram showing an arrangement of each unit of the molecular computer according to the present invention.
FIG. 3 is a block diagram showing a molecular computer according to the present invention.
FIG. 4 is a flowchart of processing of the molecular calculation method of the present invention.
FIG. 5 is a flowchart showing a modification of the processing of the molecular calculation method.
FIG. 6 is a flowchart illustrating a modified example of the processing of the molecular calculation method.
FIG. 7 is a flowchart showing a modified example of the processing of the molecular calculation method.
FIG. 8 is a flowchart showing a modified example of the processing of the molecular calculation method.
FIG. 9 is a flowchart illustrating a modification of the processing of the molecular calculation method.
FIG. 10 is a data flow diagram of a molecular calculation method.
FIG. 11 is a diagram showing an example of a procedure or function-operation reaction conversion data table.
FIG. 12 is a diagram showing an example of a calculation reaction procedure table.
FIG. 13 is a diagram showing an example of an execution procedure table.
FIG. 14 is a diagram showing an example of a control instruction procedure table.
FIG. 15 is a flowchart showing an outline of command processing;
FIG. 16 is a flowchart showing an outline of processing of a command.
FIG. 17 is a flowchart showing an outline of processing of a command.
FIG. 18 is a flowchart showing an outline of processing of a command.
FIG. 19 is a conceptual diagram showing an example of a sequence identification method.
FIG. 20 is a flowchart showing the flow of a program.
FIG. 21 is a chart showing the results of sequence identification.
FIG. 22 is a flowchart showing an encoding reaction and a decoding reaction for gene analysis.
FIG. 23 is a diagram showing an example of molecular design for gene analysis.
FIG. 24 is a schematic diagram showing the state of each molecule in a gene detection reaction step.
FIG. 25 is a schematic diagram showing the state of each molecule in a reaction system when a target is present.
FIG. 26 is a schematic diagram showing the state of each molecule in a capture step using streptavidin magnetic beads.
FIG. 27. DCN_iFIG. 4 is a schematic diagram showing the state of each molecule in the extraction step.
FIG. 28: DCN obtained in the extraction step_iSchematic diagram of amplification of a sequence complementary to.
FIG. 29 is a schematic diagram showing the state of each molecule in a step of capturing the amplification product obtained by the amplification in FIG. 28.
FIG. 30 is a schematic diagram showing the state of each molecule in a step of forming a single strand by thermal denaturation.
FIG. 31 is a schematic diagram showing the behavior of each molecule in the step of converting the information of the expressed gene into the existing molecule.
FIG. 32 is a schematic diagram showing a reaction between an inverted oligonucleotide and an existing molecule in a first step for detecting a non-expressed gene and converting it into an absent molecule.
FIG. 33 is a schematic diagram showing the state of an inverted oligonucleotide for a certain non-expressed gene.
FIG. 34 is a schematic view showing a state of a molecule in a step of extracting an inverted oligonucleotide.
FIG. 35. DCN by magnetic beads immobilized with streptavidin_kThe schematic diagram which shows the state of each molecule in the extraction process by * capture and hybridization.
FIG. 36 is a schematic view showing a nucleic acid for calculation.
FIG. 37 is a schematic diagram showing the state of each molecule in a hybridization step between a calculation nucleic acid and existing and non-existent molecules.
FIG. 38 is a schematic view showing the state of each molecule in a step of extending the molecule hybridized to the nucleic acid for operation after the step of FIG. 37.
FIG. 39: Marker oligonucleotide M₁And M₂FIG. 4 is a schematic diagram showing the state of each molecule in a detection step of a calculation result according to FIG.
FIG. 40 is a schematic view showing the state of each molecule in a step of extracting and recovering the non-existing molecule for amplifying the non-existing molecule.
FIG. 41 is a schematic diagram showing the state of each molecule in a PCR step for amplifying an absent molecule.
FIG. 42 is a schematic view showing the state of each molecule in a step of capturing an amplification product generated in the step of FIG. 41.
FIG. 43 is a schematic diagram showing the state of each molecule in the step of converting the amplification product recovered in the step of FIG. 42 into a single strand.
FIG. 44 is a schematic view showing the state of each molecule in the step of performing hybridization of an absent molecule to a single strand in the step of FIG. 43.
FIG. 45 is a schematic diagram showing a complementary nucleic acid for ligation used in the method of preparing a random library of nucleic acids for operation and a part of the nucleic acid for operation to be ligated.
FIG. 46 is a diagram showing an example of a detection device using FCS.
FIG. 47 is a view showing the analysis principle of FCS.
FIG. 48 is a diagram showing an example of a DNA capillary.
FIG. 49 is a diagram showing an example of a capillary array.
FIG. 50 is a diagram showing an example of the configuration of a detection device using a microarray in a plate.
FIG. 51 illustrates an example of a configuration of an ECA chip.
FIG. 52 is a diagram illustrating a detection principle using an ECA chip;
FIG. 53 is a diagram showing a detection method of a detection device using an electrode-immobilized DNA probe array.
FIG. 54 is a diagram showing a detection principle of a detection device using a microbead array.
FIG. 55 is a diagram showing an example of a detection device using a microtransponder.
FIG. 56 is a diagram showing an example of a detection device using a frequency conversion element.
FIG. 57 is a diagram showing an example of a detection device using surface plasmon resonance.
FIG. 58 is a diagram showing an example of a configuration of a detection device using a thread chip.
FIG. 59 is a diagram showing an example of a bead capillary array in which microbeads are arranged in a capillary.
[Explanation of symbols]
11 input unit, 12 input / output control unit, 13 storage unit, 14 operation unit, 14a translation / calculation planning and execution unit, 14b nucleic acid sequence calculation unit, 14c result analysis unit, 15 operation unit 16 storage unit 17 input / output control unit 18 input unit 19 output unit 20 output unit 21 electronic calculation unit 22 molecular calculation unit 23 and 24 communication unit 30 nucleic acid Synthesizer, 151: Automatic control unit, 152: XYZ control pipetter, 153: Thermal cycler reaction container, 154: Bead container, 155 ... Enzyme container, 156 ... Buffer solution container, 157 ... Nucleic acid container, 158 ... Detection unit, 159 ... Temperature control means, 160: transport mechanism

Claims (13)

A molecular computer that performs molecular calculation by a chemical reaction of the molecule,
A molecular control unit comprising: a reaction control unit that controls a reaction of a chemical reaction of a molecule; and an automatic detection unit that generates a chemical reaction of the molecule based on the reaction control of the reaction control unit and detects a reaction result of the chemical reaction. When,
A variable and a constant defined in association with a coding molecule obtained by encoding a molecule based on a nucleic acid sequence, and a program described by a function defined in association with the operation of the coding molecule, the operation of the coding molecule A control command generating unit that converts the control command into a control command for driving the reaction control unit that performs a reaction control of the chemical reaction of the corresponding molecule and generates a procedure of the control command.
A molecular computing device including an electronic computing unit having an output unit for outputting a procedure of the control instruction. 2. The molecular computer according to claim 1, wherein the automatic detection unit includes a fluorescence correlation analyzer that detects a reaction result of the chemical reaction using a fluorescence correlation analysis method. 2. The molecular computation device according to claim 1, wherein the automatic detection unit includes a capillary array in which capillaries on which a plurality of DNA probes are immobilized along the inner wall of the flow channel are integrated. 2. The molecular computation device according to claim 1, wherein the automatic detection unit has a microarray in which a plurality of reaction vessels are provided in an array on a plate well. 2. The molecular computer according to claim 1, wherein the automatic detection unit includes an ECA chip having an array of electrodes to which oligo DNA is immobilized. 2. The molecular computation apparatus according to claim 1, wherein the automatic detection unit includes an electrode-fixed probe array including a substrate on which a plurality of electrodes are formed and a voltage source for applying a voltage to an electric substrate. 2. The molecular computing apparatus according to claim 1, wherein the automatic detection unit includes micro beads that can be optically individually identified, and detects a reaction result of the chemical reaction using the micro beads. The molecular computer according to claim 1, wherein the automatic detection unit includes a transponder that can be individually identified, and detects a reaction result of the chemical reaction using the transponder. 2. The molecular computation device according to claim 1, wherein the automatic detection unit includes a frequency conversion element including an electrode to which the probe DNA is immobilized. The molecular computer according to claim 1, wherein the automatic detection unit includes a surface plasmon resonance detection device that detects the surface plasmon resonance using surface plasmon resonance. The molecular computer according to claim 1, wherein the automatic detection unit includes a thread chip. The said automatic detection part is the said fluorescence correlation analyzer of Claim 2, the said capillary array of Claim 3, the said microarray of Claim 4, the said ECA chip of Claim 5, and the said ECA chip. 11. The electrode-immobilized probe array according to claim 7, the microbead according to claim 7, the transponder according to claim 8, the frequency conversion element according to claim 9, and the surface plasmon resonance detection device according to claim 10. A molecular computer comprising a combination of two or more of the thread chips according to claim 11. 2. The automatic detection unit according to claim 1, wherein a bead capillary array in which a plurality of capillaries on which a plurality of optically individually identifiable microbeads are fixed along an inner wall of the flow path is integrated. Molecular computer.

Images