JP2011062085A - Apparatus for searching primer set, method and program for searching primer set - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for searching for a primer set useful for sequencing a number of target genes with little cross hybridization with non-target genes. <P>SOLUTION: The method comprises: a step in which a database is constructed from whether hybridization of the candidate primers to be included in the primer set with the target genes and plural non-target genes occurs or not; a step in which the initial primer set is generated at random; and a searching step in which the following process is repeatedly conducted: the each evaluated value of the plural primer sets formed by exchanging a primer of the primer sets with plural other candidate primers are calculated based on the rates at which the primer sets hybridize with the target genes and the non-target genes, Ct and Cn, respectively; the primer sets are subjected to stochastic selection based on the calculated evaluated value; and the data of the primer set and its evaluated value are stored in the memory device. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an apparatus and method for searching for primers, and more particularly to searching for a combination of a plurality of primers (hereinafter referred to as “primer sets”) used for gene expression level measurement by large-scale sequencing.

  In recent years, in the field of DNA sequencing, rapid scale-up has progressed, and it has become necessary to design a large-scale design using a computer for primer sequences used to identify target genes.

  By specifying a target gene group using a primer and determining the sequence, it is possible to absolute quantitate the gene expression level. Experiments that measure the expression level of a large number of genes at once are important in biology, but in order to efficiently measure the expression level of a large number of genes, multiple genes can be identified with a single primer. Primers should be used in appropriate combinations.

  Attempts to identify as many genes as possible with a small number of primers have been made in the field of PCR (for example, Non-Patent Document 1). However, the number of target genes assumed in these studies was about one hundred genes.

Pesole G. et al., “GeneUp: a program to select short PCR primer pairs that occur in multiple members of sequence lists.” Biotechniques, 1998 vol. 25, no1, pp. 112-123

  When a large number of target genes and non-target genes are targeted, the candidate sequences for primers are enormous. Examining all combinations requires unrealistic time. On the other hand, a search using a heuristic algorithm as used in Non-Patent Document 1 may fall into a local optimal solution. There is a need for an algorithm that calculates an optimal or sub-optimal combination within a realistic time.

  In addition, when cross-hybridization occurs in sequencing, a non-target gene is amplified. Therefore, it is important to minimize cross-hybridization in order to sequence many gene groups with a limited number of primers.

  The program described in Non-Patent Document 1 has an optional function for designating a small number of gene sequence lists to be excluded. By utilizing this optional function, a primer that hybridizes to an unintended gene is excluded from the search target, thereby preventing cross-hybridization. However, in this method, when there are a large number of genes to be excluded, primers to be searched are greatly limited.

  In view of the above background, the present invention provides a primer search apparatus capable of searching a large-scale primer set that can sequence many target gene groups with little cross-hybridization to non-target gene groups. The purpose is to do.

  The primer set search method of the present invention is a method for searching for a primer set that hybridizes to a target gene, the step of inputting sequences of a plurality of target genes and a plurality of non-target genes, and primers included in the primer set A search condition is input, and whether or not primer candidates that match the search condition hybridize to each of the plurality of target genes and the plurality of non-target genes is generated, and a database storing the results is generated A step of randomly selecting the number of primers to be included in the primer set from the candidate primers to generate an initial primer set, and (1) randomly selecting one primer from the primers included in the primer set. (2) Multiple selected primers For a plurality of primer sets generated when exchanged with each of the primer candidates, the ratio Ct of the target gene to hybridize and the ratio Cn of the non-target gene are determined based on the data stored in the database, (3) Based on the ratio Ct of the target gene and the ratio Cn of the non-target gene, each evaluation value of the plurality of primer sets is calculated, and (4) a probability distribution that gives a probability value corresponding to the evaluation value and the plurality of the plurality of primer sets Based on each evaluation value of the primer set, one primer set is selected from the plurality of primer sets, and (5) the selected primer set and its evaluation value are stored in the storage unit, (1) to (1) 5) a search step for repeatedly performing the process, and evaluation of the plurality of primer sets stored in the storage unit There and a step of outputting as a search result highest was primer set.

  According to the present invention, since the primer set is searched using the evaluation value calculated based on the ratio Ct of the target gene to hybridize and the ratio Cn of the non-target gene, there is little cross-hybridization to the non-target gene, In addition, combinations of primers that hybridize to many target genes can be obtained.

It is a figure which shows the structure of the primer set search apparatus of embodiment. It is a figure which shows the example of the data memorize | stored in the sequence specificity database. A dedicated circuit for calculating sequence specificity will be described. It is a figure which shows the time which calculation of sequence specificity requires. It is a figure which shows the hardware constitutions of a primer set search device. It is a figure which shows operation | movement of the primer set search apparatus of embodiment. It is a figure which shows the operation | movement of a primer combination search. It is a figure for demonstrating an evaluation value. It is a figure which shows the probability distribution for selecting a primer set.

  Hereinafter, a primer set search device and a primer set search method according to an embodiment of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram illustrating a configuration of a primer set search apparatus 10 according to the first embodiment. The primer set search device 10 is connected to the cDNA library 30. The cDNA library 30 stores the cDNA sequence in association with the characteristics of the protein expressed by the cDNA (human transcriptional gene, housekeeping gene, etc.). In the cDNA library 30, for example, data of tens of thousands of cDNAs are stored.

  The primer set search apparatus 10 has a target gene input unit 11 for inputting a target gene from the cDNA library 30 and a non-target gene input unit 12 for inputting a non-target gene. When the operator inputs the data of the characteristics of the cDNA that the operator wants to acquire as the target gene, the target gene input unit 11 reads the cDNA associated with the characteristics from the cDNA library 30. Similarly, the non-target gene input unit 12 reads out the cDNA associated with the feature from the cDNA library 30 when the operator inputs the data of the feature of the cDNA desired to be the non-target gene.

  The primer set search device 10 has a search condition input unit 13. The search condition input unit 13 receives input of the number of primers included in the primer set and the base length data of the primers. Examples of search conditions are, for example, base length: 10 nt and number of primers: 20. Based on this search condition, a search is made for a combination (primer set) of, for example, 20 primers of 10 nt that covers the target gene as much as possible and does not cover the non-target gene as much as possible.

The primer set search apparatus 10 includes a sequence specificity calculation unit 14. The sequence specificity calculation unit 14 has a function of determining whether or not all candidate primers having a base length input as search conditions are hybridized to a target gene and a non-target gene. For example, if the primer is ¹⁰ nt, there are 410 candidate primers. Whether all of these primer candidates can be hybridized with the target gene and the non-target gene is calculated by brute force, and the result is stored in a sequence specificity database (hereinafter referred to as “sequence specificity DB”) 15.

  Whether or not a primer candidate hybridizes to a gene (target gene, non-target gene) is determined by whether or not the gene contains a sequence complementary to the primer candidate sequence. If the gene contains a sequence complementary to the sequence of the candidate primer, it is determined that the candidate primer hybridizes to the gene. Even if it is not complementary to all the bases of the candidate primer sequence, if there is a sequence containing more than a predetermined percentage (for example, 80%) of complementary bases, it is determined to hybridize to that gene. Also good. The threshold for determining the degree of matching to be hybridized can be appropriately set by those skilled in the art.

  FIG. 2 shows an example of data stored in the sequence specificity DB 15. Here, for the sake of simplicity, data on whether or not candidate primers P1 to P6 hybridize to the target genes T1 to T6 and the non-target genes N1 to N6 are shown. Combinations indicated by “◯” in the figure indicate that they are hybridized. For example, the candidate primer P1 hybridizes to the target genes T1, T3, T4 and the non-target gene N1, and does not hybridize to other genes. The sequence specificity DB 15 may not store candidate primers that do not hybridize to any of the target genes T1 to T6. That is, primers that do not hybridize to any of the target genes T1 to T6 are excluded from candidate primers.

  Since the sequence specificity calculation unit 14 needs to perform sequence specificity calculation for a huge number of combinations of primer candidates and genes, a dedicated circuit is used by FPGA (Field Programmable Gate Array) or the like for high speed calculation. It may be configured. The dedicated circuit will be described later.

The primer set search apparatus 10 includes a thermodynamic characteristic calculation unit 16 that calculates Tm values of primer candidates. The Tm value is the temperature at which half of the primers hybridize to the target gene. The thermodynamic characteristic calculation unit 16 first calculates the Gibbs free energy change ΔG in the hybridization reaction from the primer sequence and the target sequence using the nearest base pair method (Nearest Neighbor method). At this time, the thermodynamic characteristic calculation unit 16 may consider the ion concentration of Mg ²⁺ , Na ⁺ or the like.

Next, the hybridization rate f at temperature T is determined using the following equation (1).

  Here, the primer concentration Cp and the gas constant R are set with conditions for actual sequencing using primers. For example, the primer concentration Cp can be 0.5 (μmol / l) and the gas constant R can be 1.9872 (cal / mol K). The thermodynamic characteristic calculation unit 16 obtains T at which f = 0.5 and sets it as the Tm value.

  The primer set search apparatus 10 has a temperature condition input unit 17 that inputs a temperature Tc as a temperature condition using the primer set. The temperature condition input unit 17 receives input of an upper limit value and a lower limit value of the temperature condition Tc. In addition, the primer set search apparatus 10 includes a database update unit (hereinafter referred to as “DB update unit”) 18 that updates the sequence specificity DB 15 by excluding primer candidates having Tm values that do not match the temperature condition Tc. .

  The primer set search device 10 uses the data stored in the sequence specificity DB 15 to hybridize as much as possible to the target gene and to search for a primer set with as little cross hybridization to the non-target gene as possible. have. The primer set search algorithm will be described later.

  The primer set search apparatus 10 has an output unit 20 that outputs the primer set searched by the primer set search unit 19. The output unit 20 may display the searched primer set, or may output the search result from another device by transmitting the search result to the other device.

  With reference to FIG. 3, the dedicated circuit constituting the sequence specificity calculation unit 14 will be described. FIG. 3 shows an example in which a b-base length primer candidate is compared with a target gene and a non-target gene. In FIG. 3, the complementary sequence of the candidate primer is described as “query sequence”, and the target gene and non-target gene are described as “database sequence”.

  The dedicated circuit has a sequence comparison unit 40 that compares whether the sequences having the base length b (nt) match each other. The bases contained in the sequences to be compared are compared in parallel. In FIG. 3, two sequences having a base length b (nt) are input as comparison targets, and b bases included in the sequences are simultaneously compared.

  A threshold value is input to the sequence comparison unit 40. The sequence comparison unit 40 determines that a hit occurs when the sequence matching rate is higher than the threshold. For example, when “10” is input as the threshold value, a hit is determined if 10 or more of the bases in the sequence match. Note that the threshold value input to the sequence comparison unit 40 is calculated from the base length input as the search condition and the matching rate. The sequence comparison unit 40 outputs “hit” information based on the comparison result. The value of the “hit” information is, for example, “1” when the number of matched bases is greater than or equal to a threshold, and “0” when less than the threshold.

  The dedicated circuit constitutes a 32 parallel arrangement comparison circuit 41 by collecting 32 arrangement comparison units 40 described above. The 32-parallel sequence comparison circuit 41 compares the b-base length sequence with the (b + 31) base-length sequence and outputs 32-bit information (“hit”). The dedicated circuit further has two parallel comparison circuits 41 in parallel. With the above configuration, the dedicated circuit can perform sequence specificity calculation with high throughput.

  In order to further increase the throughput of sequence specificity calculation, the target gene sequence is stored in a shift-type register that shifts by 2 bases in one clock, and only the lower bits of the register to be calculated are connected to the comparison circuit. But you can. Normally, primer candidate sequences stored in a register are used as an input to the comparison circuit. However, in order to calculate a large number of primer candidates, the contents of the register must be exchanged. With the above-described structure, the input to the comparison circuit can be switched at high speed with a small number of logical resources, and the throughput of the 32 parallel array comparison circuits 41 connected downstream can be maximized.

  FIG. 4 is a diagram showing the calculation time of sequence specificity (alignment). The experimental condition was that 488 candidate primer sequences (25 nt) at the 3 ′ end of the human ERBB gene were aligned with the human mRNA database without gaps, and the time required to calculate the hit threshold as 80% was measured (CPU : Intel Xeon 3.8 GHz, OS: Linux). In the case of Naive implemented in C language, it took more than 10,000 seconds, and even using the high-speed alignment software LAST, it took nearly 1000 seconds. In a system in which the dedicated circuit is configured in the FPGA, alignment was possible in just over 10 seconds. Heretofore, the dedicated circuit constituting the sequence specificity calculation unit 14 has been described.

  FIG. 5 is a diagram showing a hardware configuration of the primer set search apparatus 10 described above. The primer set search apparatus 10 is configured by a computer in which a CPU 50, a RAM 51, a ROM 52, a communication interface 54, a hard disk 55, an operation unit 56, and a display 57 are connected by a data bus 59. Further, the FPGA board 58 having a dedicated circuit for calculating the sequence specificity is mounted on the computer. The function of the primer set search device 10 described above is realized by the CPU 50 executing arithmetic processing according to the program 53 written in the ROM 52. Such a program 53 is included in the scope of the present invention. The hard disk 55 stores data of the sequence specificity DB 15.

  FIG. 6 is a diagram illustrating the operation of the primer set search apparatus 10 of the present embodiment. First, the primer set search device 10 receives input of a target gene sequence and a non-target gene sequence (S10). Specifically, when the operator specifies the characteristics of the target gene and the characteristics of the non-target gene, the primer set search apparatus 10 reads out the target gene sequence and the non-target gene sequence that match the characteristics from the cDNA library 30.

  Next, the primer set search apparatus 10 receives input of the base length of the primer included in the primer set, the number of primers included in the primer set, and the temperature condition Tc as search conditions (S12). The primer set search apparatus 10 treats all primers having the input base length as candidate primers to be included in the primer set.

  Subsequently, the primer set search apparatus 10 calculates whether or not each candidate primer hybridizes to the target gene and the non-target gene, and based on the calculation result, the sequence specificity DB 15 as shown in FIG. Generate (S14). Here, when the target gene sequence includes a complementary sequence that matches the primer sequence at a ratio equal to or higher than a predetermined threshold value, it is determined to hybridize. This threshold value can be appropriately set depending on the system design. In the generation of the sequence specificity DB15, primers that do not hybridize to any target gene may be excluded from the primer candidates and not included in the sequence specificity DB15.

  The primer set search apparatus 10 obtains the Tm value of the candidate primer stored in the sequence specificity DB 15 using the nearest base pair method (Nearest Neighbor method). Subsequently, the data of the sequence specificity DB 15 is updated based on the obtained Tm value and the input temperature condition Tc (S16). Specifically, primer candidates having Tm values that do not match the temperature condition Tc are deleted from the sequence specificity DB 15. By excluding such primer candidates from the sequence specificity DB 15 in advance, it is possible to reduce the calculation load for the primer set search.

  Next, the primer set search device 10 searches for primer combinations to be adopted for the primer set from the primer candidates based on the data stored in the sequence specificity DB 15 (S18).

  FIG. 7 is a diagram illustrating an operation of searching for a combination of primers. The primer set search unit 19 randomly selects the same number of primer candidates as the number of primers input as a search condition, and generates an initial primer set (S30). Here, when selecting candidate primers, it may be allowed to select the same candidate candidates in duplicate.

Next, determine the evaluation value _{S 0} of the initial primer set (S32). Hereinafter, when the evaluation value is referred to without specifying the primer set, it is referred to as “evaluation value S”. The evaluation value S takes a higher value as the proportion Ct of the target gene to hybridize increases, and takes a lower value as the proportion Cn of the non-target gene to cross hybridize increases.

  FIG. 8 is a diagram for explaining the evaluation value S. FIG. 8 shows an example in which the evaluation value S is obtained for a primer set composed of a combination of primers P1, P3, and P4 among the primer candidates P1 to P6.

  The primer set shown in FIG. 8 covers T1 to T5 among the target genes T1 to T6. That is, primer P1 hybridizes to target genes T1, T3, T4, primer P3 hybridizes to target genes T2, T4, and primer P4 hybridizes to target genes T1, T5. Therefore, the ratio Ct of the target gene covered by the primer set is 5/6. Further, since the primer set shown in FIG. 8 cross-hybridizes to non-target genes N1, N3, and N4, the ratio Cn of non-target genes that cross-hybridize is 3/6. Using the data stored in the sequence specificity DB 15 as described above, the ratio Ct of the target gene to which the primer set hybridizes and the ratio Cn of the non-target gene are obtained.

The evaluation value S is obtained by the following formula (2) or formula (3) using the ratio Ct of the target gene and the ratio Cn of the non-target gene, for example.
In addition, the said formula is an illustration and the evaluation value S can also be determined using another formula.

  Next, the primer set search unit 19 randomly selects one primer from the primer set (S34). For convenience of explanation, the selected primer is referred to as primer A. The primer set search unit 19 selects a primer to be exchanged with the primer A from the primer candidates (S36).

  Specifically, the primer set search unit 19 includes a primer other than the primer A included in the primer set (for example, if the number of primers is c, (c-1) primers other than the primer A) and a dimer. Primers that do not form are selected as exchangeable primer candidates. Here, for convenience of explanation, it is assumed that there are m primer candidates to be exchanged.

And the evaluation value S _t of the primer sets of before replacing the primer A. t indicates the number of times of repeated processing for exchanging the primer A included in the primer set with another primer candidate. In the case of the initial primer set, t = 0. The primer set search apparatus 10 calculates the evaluation values S _{t + 1,1} , S _{t + 1,2} ,... S _{t + 1, m} of _{the m} primer sets after being replaced with each of the m primer candidates ( S38).

Next, the primer set search unit 19 determines a candidate primer to be exchanged with the primer A based on the evaluation values S _{t + 1,1} , S _{t + 1,2} ,... S _{t + 1, m} and determines a primer set. Specifically, first, based on the probability distribution that takes a larger probability value as the evaluation value S is larger, and the calculated evaluation values S _{t + 1,1} , S _{t + 1} , ₂ ,... S _{t + 1, m of} the primer set. Then, a probability value for selecting each primer set is determined (S40).

FIG. 9 is a diagram showing a probability distribution. The probability distribution shown in FIG.
Is a graph. In equation (4), min [a, b] is a function that selects the smaller of a and b. As described above, when the evaluation value is higher than the present value by using the smaller value by comparing 1 and exp {k (S _{t + 1, i} −S _t )} using the min function. , All become 1, and are selected with equal probability. In addition, you may obtain | require probability distribution by following formula (5), for example, without using min function.

The primer set search unit 19 does not always select a primer set having the maximum evaluation value S, but selects a primer set using a probability distribution. In the example shown in FIG. 9, the probability value that the primer set of the evaluation value S _{t + 1,1} is selected is P1, the probability value of the primer set of the evaluation value S _{t + 1,2} is P2, and the evaluation value S _The probability value that the primer set of _{t + 1,3} is selected is P3. The primer set search unit 19 determines a primer set according to the probability value determined by the evaluation value S, and stores the evaluation value S of the primer set in the storage unit (S42).

  The primer set search unit 19 determines whether or not the primer set search end condition is satisfied (S44). As the termination condition, for example, the number of primer set searches, that is, the number of primer replacements, can reach a predetermined threshold value n. As the threshold value n, for example, an order value of 10,000 times can be set. A primer set close to an ideal solution that maximizes the number of target genes that hybridize under given search conditions and minimizes cross hybridization can be obtained.

  When the termination condition is satisfied, the primer set search unit 19 determines a primer set having the maximum evaluation value S stored in the storage unit as a search result (S46).

  Returning to FIG. 6, the primer set search apparatus 10 outputs the primer set searched by the primer set search unit 19 as a search result (S20). Heretofore, the configuration and operation of the primer set search device 10 of the first embodiment have been described.

  The primer set search apparatus 10 and the primer set search method according to the present embodiment evaluate the primer set obtained based on the ratio Ct of the primer hybridizing to the target gene and the ratio Cn of the primer cross-hybridizing to the non-target gene. Since the primer set is searched using the value S, it is possible to obtain a primer set that minimizes cross-hybridization and hybridizes to as many target genes as possible.

  In addition, the primer set search apparatus 10 and the primer set search method select a candidate primer to be exchanged based on a probability distribution that takes a higher probability value as the evaluation value becomes higher and an evaluation value of the primer set, so that a local maximum is reached. It is possible to search for an appropriate primer set.

(Second Embodiment)
Next, the primer set search device 10 and the primer set search method according to the second embodiment of the present invention will be described. The basic configuration of the primer set search device 10 of the second embodiment is the same as that of the first embodiment (see FIG. 1), but the primer set search device 10 of the second embodiment is The difference is that the probability distribution for selecting a primer set is changed when searching for the primer set.

In the second embodiment, as the evaluation value S,
, And the probability Pi when replacing candidate primer A with another candidate primer,
Ask for.

In the second embodiment, for each iterative process of exchanging primer, and the evaluation value S _t primer sets obtained after iteration evaluation value S _{t + 1} and t times primer sets obtained after t + 1 iterations process In comparison, if S _{t + 1} <S _t , k _i is multiplied by 1.05 to change the probability distribution. By changing the probability distribution in this way, when the evaluation value S _{i + 1} becomes larger than the previous evaluation value S _i , the probability that a primer set having a high evaluation value S is selected is increased by increasing k _i. Make it high. As a result, it is possible to quickly reach a primer set close to an ideal solution that maximizes the number of target genes that hybridize under a given search condition and minimizes cross hybridization.

In the present embodiment, the example in which the probability distribution is changed based on the magnitude relationship between the evaluation values S _t and S _{t + 1} of the primer set has been described. However, the probability distribution may be changed according to another criterion. For example, the probability distribution may be changed based on the number of times that the repetition process for exchanging the primer A is performed.

Next, examples of the present invention will be described.
First, experimental conditions will be described. 1000 target gene sequences were randomly selected, and 1000 non-target gene sequences were randomly selected from housekeeping genes. Search conditions are as shown in Table 1 below.

Under this search condition, the search was repeated 20,000 times. For comparison, a primer set was searched by the Greedy algorithm. Table 2 below shows the number of target gene covers and the number of cross-hybridizations to non-target genes by the searched primer sets. In the following table, the values in parentheses are values obtained by subtracting the number of non-target genes from the number of target genes. It can be evaluated that the larger the numerical value, the less the cross-hybridization and the primer set that hybridizes to many target genes.

  According to the results shown in Table 2 above, subtracting the number of cross-hybridization from the target gene cover number to the non-target gene and evaluating each experimental result, the present invention is It can be seen that an appropriate search result was obtained from the Greedy algorithm.

  As mentioned above, although the primer set search apparatus and primer set search method of the present invention have been described in detail with reference to the drawings, the present invention is not limited to the above-described embodiments.

  In the above-described embodiment, the primer set search end condition is that the search count has reached a preset threshold, but the primer set search end condition is not limited to this. For example, the end condition may be that the fluctuation width of the evaluation value for each repetition process has converged within a certain range, or that the evaluation value has exceeded a predetermined threshold.

  In the above-described embodiment, an example in which the Tm value of the primer candidate is calculated by the thermodynamic characteristic calculation unit 16 has been described. However, the Tm value of the primer candidate may be stored in advance. Alternatively, the Tm value obtained once may be accumulated for the next and subsequent searches. Thereby, the time for calculating the Tm value of the primer candidate can be omitted or shortened, and an appropriate primer set can be searched for in a shorter time.

  In the above-described embodiment, the example in which the calculation of the thermodynamic characteristics is performed and the primer candidates are narrowed down after the calculation of the sequence specificity has been described. However, the thermodynamic characteristics are calculated first and the temperature condition Tc is calculated. Based on the above, candidate primers may be narrowed down. In particular, when the thermodynamic characteristics of the primer candidate are obtained in advance, it is effective to narrow down the temperature condition Tc first.

  INDUSTRIAL APPLICABILITY The present invention is useful as a primer search device that searches for primer sets used for sequencing, and can search for primer sets that hybridize to many target gene groups with little cross-hybridization to non-target gene groups. .

DESCRIPTION OF SYMBOLS 10 Primer set search apparatus 11 Target gene input part 12 Non-target gene input part 13 Search condition input part 14 Sequence specificity calculation part 15 Sequence specificity DB
16 thermodynamic characteristic calculation unit 17 temperature condition input unit 18 DB update unit 19 primer set search unit 20 output unit 30 cDNA library 40 sequence comparison unit 41 32 parallel sequence comparison circuit 50 CPU
51 RAM
52 ROM
53 Program 54 Communication Interface 55 Hard Disk 56 Operation Unit 57 Display 58 FPGA Board 59 Data Bus

Claims (15)

A method for searching for a primer set that hybridizes to a target gene,
Inputting sequences of a plurality of target genes and a plurality of non-target genes;
Entering search conditions for primers to be included in the primer set;
Determining whether or not primer candidates that match the search conditions hybridize to each of the plurality of target genes and the plurality of non-target genes, and generating a database storing the results;
Randomly selecting a number of primers to be included in the primer set from the primer candidates to generate an initial primer set;
(1) The purpose of hybridizing a plurality of primer sets generated when one primer is randomly selected from the primers included in the primer set, and (2) the selected primer is replaced with each of a plurality of candidate primers. Gene ratio Ct and non-target gene ratio Cn are obtained based on the data stored in the database, and (3) a plurality of primer sets based on the target gene ratio Ct and the non-target gene ratio Cn (4) One primer set from the plurality of primer sets based on a probability distribution that gives a probability value corresponding to the evaluation value and each evaluation value of the plurality of primer sets (5) Store the selected primer set and its evaluation value in the storage unit. A search step of performing repeatedly the process of serial (1) to (5),
Outputting a primer set having the highest evaluation value among a plurality of primer sets stored in the storage unit as a search result;
A primer set search method comprising:   2. The primer set search according to claim 1, wherein the step of generating the database excludes primers that do not hybridize to any of the target genes from primers that match the search conditions and does not include them in the database. Method.   The primer set search method according to claim 1 or 2, wherein the evaluation value of the primer set is higher as the ratio Ct of the target gene is higher and lower as the ratio Cn of the non-target gene is higher. The evaluation value S of the primer set is
S = Ct−α × Cn (where α> 0)
The primer set search method according to claim 3, which is calculated by: The evaluation value S of the primer set is
S = (Ct + α) / (Cn + β) (where β ≠ 0)
The primer set search method according to claim 3, which is calculated by:   The primer set search method according to any one of claims 1 to 5, wherein the probability distribution is changed based on a change in an evaluation value stored in the storage unit.   The primer set search method according to any one of claims 1 to 5, wherein the probability distribution is changed based on the number of repetitions of the processes (1) to (5). Calculating a Tm value for each primer candidate and storing the Tm value in the database;
Entering temperature conditions;
Deleting the candidate primer having a Tm value that does not match the input temperature condition from the database;
The primer set search method according to any one of claims 1 to 7.   The primer search method according to claim 1, wherein the search step selects a candidate primer that does not form a dimer with a primer contained in the primer set, as a candidate primer for exchanging one primer.   The step of generating the database includes obtaining the target gene or the non-target gene having a sequence portion that matches a complementary base sequence of a candidate primer at a ratio equal to or higher than a predetermined threshold as a gene to which the candidate primer hybridizes. Item 10. The primer set search method according to any one of Items 1 to 9.   The step of creating the database calculates whether a candidate primer hybridizes to the target gene or the non-target gene by using a dedicated circuit generated by an FPGA (Field Programmable Gate Array). The primer set search method in any one of -10.   The primer set search method according to claim 1, wherein the search step is ended when the number of repetitions of the processes (1) to (5) reaches a predetermined threshold value.   The primer set search method according to claim 1, wherein the search step ends when an evaluation value of the primer set stored in the storage unit reaches a predetermined threshold value. A program for searching for a primer set that hybridizes to a target gene,
Inputting sequences of a plurality of target genes and a plurality of non-target genes;
Entering search conditions for primers to be included in the primer set;
Determining whether or not primer candidates that match the search conditions hybridize to each of the plurality of target genes and the plurality of non-target genes, and generating a database storing the results;
Randomly selecting a number of primers to be included in the primer set from the primer candidates to generate an initial primer set;
(1) The purpose of hybridizing a plurality of primer sets generated when one primer is randomly selected from the primers included in the primer set, and (2) the selected primer is replaced with each of a plurality of candidate primers. Gene ratio Ct and non-target gene ratio Cn are obtained based on the data stored in the database, and (3) a plurality of primer sets based on the target gene ratio Ct and the non-target gene ratio Cn (4) One primer set from the plurality of primer sets based on a probability distribution that gives a probability value corresponding to the evaluation value and each evaluation value of the plurality of primer sets (5) Store the selected primer set and its evaluation value in the storage unit. And a step to repeat the processing of the serial (1) to (5),
Outputting a primer set having the highest evaluation value among a plurality of primer sets stored in the storage unit as a search result;
A program that executes An apparatus for searching for a primer set that hybridizes to a target gene,
Means for inputting sequences of a plurality of target genes and a plurality of non-target genes;
Means for inputting search conditions for primers to be included in the primer set;
Means for determining whether or not primer candidates that match the search conditions hybridize to each of the plurality of target genes and the plurality of non-target genes, and generating a database storing the results;
Means for randomly selecting the number of primers to be included in the primer set from the candidate primers and generating an initial primer set;
(1) The purpose of hybridizing a plurality of primer sets generated when one primer is randomly selected from the primers included in the primer set, and (2) the selected primer is replaced with each of a plurality of candidate primers. Gene ratio Ct and non-target gene ratio Cn are obtained based on the data stored in the database, and (3) a plurality of primer sets based on the target gene ratio Ct and the non-target gene ratio Cn (4) One primer set from the plurality of primer sets based on a probability distribution that gives a probability value corresponding to the evaluation value and each evaluation value of the plurality of primer sets (5) Store the selected primer set and its evaluation value in the storage unit. Serial (1) and the means to repeat the process to (5),
Means for outputting a primer set having the highest evaluation value among a plurality of primer sets stored in the storage unit as a search result;
A primer set search device comprising: