JP5344670B2 - Gene expression analysis method, gene expression analysis apparatus, and gene expression analysis program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for easily comparing reference profile data (reference PD) with created measurement profile data (measurement PD) and analyzing an expression state of a gene. <P>SOLUTION: The method includes a reference PD acquisition step S1 of showing, as a first waveform, a reference range of values equivalent to the number of bases based on positions and a detection amount of the values equivalent to the number of bases of DNA fragments, identifying and storing transcript seeds from which a predetermined peak in the first waveform comes, and acquiring the stored reference PD; a measurement PD creation step S2 of creating a measurement PD showing, as a second waveform, a measurement range of values equivalent to the number of bases obtained on the basis of positions and a detection amount of values equivalent to the number of bases of DNA fragments to be measured; a peak association step S3 applying correction processing and associating positions of peaks of a part of or all of at least either the first waveform or the second waveform; and a gene expression analyzing step S4 of reading information of the derivation of associated peak from transcript seed information, identifying the gene of the peak of the measurement PD, and thus, analyzing the expression state of the gene. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to a gene expression analysis method, a gene expression analysis apparatus, and a gene expression analysis apparatus that analyze a gene expression state using profile data represented as a single waveform having a plurality of peaks derived from a plurality of expressed gene transcripts, and The present invention relates to a gene expression analysis program.

  Genomic science is moving to so-called post-sequence research, which is the elucidation of the control mechanism related to gene expression and the elucidation of the function of gene products after the symbolic event of "complete decoding of human genetic information (human genome)". Advancing post-sequence research can elucidate various life phenomena, so it is not only highly scientifically significant, but also contributes greatly to the development of pharmaceuticals and the possibility of realizing advanced custom-made treatments, etc. There is. Therefore, the degree of expectation for the progress of post-sequence research is very high.

An example of such post-sequence studies is gene expression analysis that analyzes gene expression status. Even in the same gene, its expression level changes every moment depending on the state of the living body, and a plurality of mRNAs corresponding to different proteins are generated from one gene by alternative splicing or the like. In higher animals and plants, non-coding RNAs and microRNAs that are not translated into proteins are also produced, and it is clear that these are regulated in gene expression and preserved in various species. It is becoming.
Among these RNAs, the analysis of gene expression by analyzing the expression level of mRNA that is directly related to the expression of protein, as described above, is particularly effective for realizing more advanced custom-made treatments and the like. is important.

  As a method for performing gene expression analysis, a differential display method, a SAGE (Serial Analysis of Gene Expression) method, a DNA microarray method, a DNA chip method, etc. that can comprehensively analyze the expression state of a gene are widely used. ing.

Also, in recent years, as a method for performing gene expression analysis, a high coverage gene profiling method (hereinafter referred to as “HiCEP method”) that enables comprehensive and more accurate gene expression analysis. )) Has been developed and attracts attention (see, for example, Patent Document 1).
In the HiCEP method, a cDNA obtained by reverse transcription of mRNA is cleaved with two types of restriction enzymes, and a DNA fragment of about 20 bases called an adapter having a special base sequence is ligated thereto, and the special base described above is further ligated. Capillaries that perform selective PCR using fluorescently labeled selective PCR primers having a base sequence that is complementary to the adapter having the sequence, which are separated according to various lengths by capillary electrophoresis and analyzed Measurement profile data expressed as a single waveform having a plurality of peaks is obtained using a DNA sequencer. In this way, the plurality of peaks described above has about 200 peaks in one waveform. In principle, peaks derived from transcripts of the same gene are detected at the same migration position (peak position on the profile) even if the intensity changes due to a difference in conditions for the same sample.

  Therefore, the HiCEP method has the advantage that the expression state of an unknown gene whose base sequence has not been determined can be analyzed, and the total transcription product (total mRNA) of the expressed gene can be analyzed. If the ratio of the transcript (mRNA) to be analyzed is defined as the coverage ratio, the coverage ratio of the conventional method described above is 10 to 30%, whereas the HiCEP method has a coverage ratio of 70 to 80%. Have achieved. Furthermore, it is possible to reliably capture a minute fluctuation amount of about ± 20%. Thus, the HiCEP method achieves high accuracy and high sensitivity that could not be realized by the conventional DNA microarray method or the like.

International Publication No. 02/048352 Pamphlet

  However, the differential display method, the SAGE (Serial Analysis of Gene Expression) method, the DNA microarray method, the DNA chip method, etc. can only deal with genes whose base sequences are known in advance, and have low sensitivity (for example, for detection). It is said that the required amount of mRNA variation is 2 to 3 times.), And the reproducibility of the results cannot be said to be sufficient unless the expression level varies greatly.

  In addition, the HiCEP method described in Patent Document 1 can analyze the expression state of a gene transcript with high coverage, high accuracy, and high sensitivity that could not be achieved by conventional methods, but it uses electrophoresis. Therefore, the peak may appear at a position different from the original base number due to the same size of the cDNA fragment or the three-dimensional molecular structure of the cDNA fragment during electrophoresis. In addition, the peak position of the measurement profile data obtained due to deterioration of the electrophoresis polymer used in the capillary DNA sequencer, use of a polymer from a different production lot, or change in voltage or temperature during electrophoresis is shifted. May end up. Also, when using sequencers of different models, the position of several bp peaks may be measured differently due to differences in measurement conditions such as different polymers. Care must be taken when comparing data from different researchers. is there.

  Therefore, when the created measurement profile data is compared with the reference profile data used as a reference, or when measurement profile data created under the same conditions on different facilities or on different days are compared, the peak In some cases, the positions of these do not match, making comparison difficult. In such a case, for example, each peak is subjected to correction processing such as multiplication by a correction coefficient related to changes in temperature or voltage during electrophoresis, and the peak appearance position and the like are corrected, and then these are compared. Although it is conceivable, it is very burdensome and time consuming to perform such processing on all the obtained peaks. In addition, unlike the number of bases of the fragment sequence constituting the peak, an absolute peak size serving as a reference cannot be assumed essentially.

  The present invention has been made in view of the above problems, and can easily compare the created measurement profile data with the reference profile data, or can compare the created measurement profile data with each other, and can express the gene. It is an object of the present invention to provide a gene expression analysis method, a gene expression analysis apparatus, and a gene expression analysis program that can easily analyze a state.

(1) The gene expression analysis method according to the present invention that solves the above-described problem is characterized in that the expression state of a gene is determined using profile data that represents a plurality of peaks derived from a plurality of expressed gene transcripts as one waveform. A gene expression analysis method for analysis, which includes a reference profile data acquisition step, a measurement profile data creation step, a peak association step, and a gene expression analysis step.

That is, in the gene expression analysis method according to the present invention, in the reference profile data acquisition step, the position corresponding to the number of bases of a DNA fragment obtained by amplifying a part of cDNA reverse-transcribed from a plurality of gene transcripts, A reference range of a base number equivalent value obtained based on a detected amount equivalent to the transcription amount of the gene transcript at the position is represented as a first waveform, and the transcript type information of the gene transcript as the first waveform Reference profile data that identifies and stores a predetermined peak and the transcript type from which the peak is derived is acquired in advance, and then a plurality of gene transcripts are reverse-transcribed in a measurement profile data creation step. A position corresponding to the number of bases of a DNA fragment to be measured obtained by amplifying a part of the obtained cDNA, and the measurement pair at the position Creating a measurement profile data representing the measuring range of the number of bases corresponding value obtained on the basis of the detected amount of the transfer amount equivalent of goods as the second waveform. Next, in the peak association step, correction processing is performed to correct a partial or entire region of at least one of the first waveform and the second waveform and adjust the position of the peak, and the second waveform The plurality of peaks including the region of interest in the first waveform and the plurality of peaks in the first waveform are associated with each other to thereby match the plurality of peaks including the region of interest and the first waveform corresponding to the plurality of peaks. In the gene expression analysis step, the information on the gene transcript derived from the peak of the associated measurement profile data is read from the transcript species information and correlated. The gene expression state is analyzed by specifying the gene from which the peak of the measurement profile data is derived.
In the peak matching step of the present invention, the peak position of the reference profile data acquired in the reference profile data acquisition step and the peak position of the measurement profile data created in the measurement profile data creation step Are matched, the corresponding peak of the reference profile data is associated with the peak of the measurement profile data, and the peak position of the reference profile data acquired in the reference profile data acquisition step and the measurement profile data creation step If the position of the peak of the measurement profile data created in step 1 is partly or completely shifted, at least one of these waveforms is partially or entirely so that the similarity between these waveforms is the highest. Correct the area and adjust the peak position After performing the correction processing of integer associates the peak measured profile data and the peak of the reference profile data.

Therefore, in the gene expression analysis method according to the present invention, the acquired reference profile data and the created measurement profile data are represented as waveforms and compared, so that the peak of the reference profile data and the peak of the created measurement profile data are compared. Can be easily associated. And the gene expression state can be analyzed appropriately by specifying the gene that is derived from the peak of interest in the measurement profile data that can be correlated.
In addition, if the peak position of the reference profile data and the peak position of the measurement profile data match, these peak positions are matched as they are without correction processing, and only when these peak positions are shifted. Since the correction processing for adjusting the position of the peak is performed for association, quick association is possible.

(2) Whether the reference profile data acquisition step in the present invention acquires the reference profile data from a database storing known profile data or artificially creates it from the transcript type information Acquired by adding or deleting one or more peaks from the known profile data or the measurement profile data, or by combining a plurality of the reference profile data, or the measurement profile data It is preferable to obtain by combining a plurality.
In this way, reference profile data can be acquired quickly and easily and compared with measurement profile data.

( 3 ) In this invention, it is preferable that the correction process in the said peak matching process is performed by the function approximation based on a Gaussian function.
In this way, the correction process can be performed with high accuracy and simplicity while reducing the burden on the correction process.

( 4 ) In the present invention, the correction processing in the peak association step moves the peak position of the measurement profile data based on the peak position of the reference profile data, or the peak position of the measurement profile data The peak position of the reference profile data is moved on the basis of the above, or both the peak position of the reference profile data and the peak position of the measurement profile data are moved, and the peak of the measurement profile data and the peak It is preferable to associate the peak of the reference profile data.
In this way, it is possible to appropriately associate the reference profile data peak with the measurement profile data peak.

( 5 ) The gene expression analysis step according to the present invention is displayed so that a peak in which the peak of the reference profile data and the peak of the measurement profile data can be correlated can be distinguished from a peak that cannot be correlated. In addition, when gene information regarding the gene from which the peak is derived is added to the peak of the reference profile data, the gene of the peak associated in the measurement profile data is obtained by quoting the gene information. It is preferable to identify and analyze the expression state of the gene.

  In this way, the obtained analysis result can easily identify the peak that could not be matched, and the peak of the measurement profile data that can be matched is identified by the gene from which it was derived. Therefore, for example, when this is referred to later, it is highly useful to the user, such as what gene the peak is derived from, and it is not necessary to reanalyze the base sequence.

( 6 ) It is preferable that the gene expression analysis step in the present invention includes a step of adding related information regarding the peak with respect to a peak that could not be matched in the peak matching step, ( 7 ) The related information includes at least one of an evaluation value based on a correlation coefficient related to the similarity of the waveform, a peak position, a primer set, an expression intensity, a peak shape characteristic, and sample cell information and experimental information. It is preferable.

  If it does in this way, since related information about the peak can be added in the gene expression analysis step for the peak that could not be matched in the above-described peak matching step, for example, when this is referred to later, It is possible to obtain evaluation values based on correlation coefficients related to waveform similarity, peak positions, primer sets, expression intensity, peak shape characteristics, and related information such as sample cell information and experimental information. Therefore, the analysis result obtained in this way has a higher utility value for the user.

( 8 ) The gene expression analysis apparatus according to the present invention analyzes a gene expression state using profile data expressed as one waveform having a plurality of peaks derived from a plurality of expressed gene transcripts. An analysis device, a position corresponding to the number of bases of a DNA fragment obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcription products, and a detection amount corresponding to the transcription amount of the gene transcription product at that position The reference range of the base number equivalent value obtained based on the above is represented as a first waveform, and as the transcript type information of the gene transcript, the predetermined peak in the first waveform and its peak are derived The reference profile data acquisition means for acquiring the reference profile data that has been identified and stored in advance and the plurality of gene transcripts are reversed. A base obtained on the basis of the position corresponding to the number of bases of a DNA fragment that is a measurement object obtained by amplifying a part of the copied cDNA and the detection amount corresponding to the transcription amount of the measurement object at that position Measurement profile data creating means for creating measurement profile data representing a measurement range of a number equivalent value as a second waveform, and a part or all of at least one of the first waveform and the second waveform Is corrected and the peak position is adjusted, and the plurality of peaks including the region of interest in the second waveform and the plurality of peaks in the first waveform are associated with each other. Peak associating means for associating a plurality of peaks including the region with a plurality of peaks in the first waveform corresponding to the plurality of peaks, and the associated measurement profile Gene expression that analyzes gene expression status by reading the gene transcript information from the peak of profile data from the transcript species information and identifying the gene from which the peak of the associated measurement profile data is derived have a analysis means, said peak correlating means includes a position of the peak of the reference profile data obtained by the reference profile data acquisition means, the measurement of the profile data peak created by the measurement profile data creation means If the position matches, the corresponding peak of the reference profile data is associated with the peak of the measurement profile data, and the peak position of the reference profile data acquired by the reference profile data acquisition means In the measurement profile data creation means When the position of the peak of the measured profile data is partially or entirely deviated, at least one of these waveforms is partially or entirely so that the similarity between these waveforms is the highest. the area after performing the correction process of adjusting the position of the corrected peak, and the peak of the reference profile data, the peak of the measured profile data, characterized Rukoto associates.

In this way, since the acquired reference profile data and the created measurement profile data are represented as waveforms, they can be easily compared, and the correspondence between the peak of the reference profile data and the peak of the measurement profile data is also possible. And analysis of gene expression status can be easily performed.
Further, in this way, when the peak position of the reference profile data and the peak position of the measurement profile data match, these peak positions are directly associated without correction processing, and Only when the position is shifted, the correction processing for adjusting the position of the peak is performed for the association, so that the association can be performed quickly.

( 9 ) Whether the reference profile data acquisition means in the present invention acquires the reference profile data from a database storing known profile data, or artificially creates it from the transcript type information. Acquired by adding or deleting one or more peaks from the known profile data or the measurement profile data, or by combining a plurality of the reference profile data, or the measurement profile data Preferably, it is obtained by synthesizing using a plurality.
In this way, reference profile data can be acquired quickly and easily and compared with measurement profile data.

(1 0 ) In the present invention, it is preferable that the correction processing in the peak association unit is performed by function approximation based on a Gaussian function.
In this way, the correction process can be performed with high accuracy and simplicity while reducing the burden on the correction process.

(1 1 ) In the present invention, the correction processing in the peak association unit moves the peak position of the measurement profile data based on the peak position of the reference profile data, or the peak of the measurement profile data The peak position of the reference profile data is moved with respect to the position, or both the peak position of the reference profile data and the peak position of the measurement profile data are moved to It is preferable to associate the peak of the reference profile data.
In this way, it is possible to appropriately associate the reference profile data peak with the measurement profile data peak.

(1 2 ) The gene expression analyzing means in the present invention is capable of distinguishing between a peak in which the peak of the reference profile data and the peak of the measurement profile data can be associated with a peak that cannot be associated. And when the gene information related to the gene from which the peak is derived is added to the peak of the reference profile data, the gene of the peak associated in the measurement profile data by quoting the gene information It is preferable to identify the gene and analyze the expression state of the gene.

  In this way, the obtained analysis result can easily identify the peak that could not be matched, and the peak of the measurement profile data that can be matched is identified by the gene from which it was derived. Therefore, for example, when this is referred to later, it is highly useful to the user, such as what gene the peak is derived from, and it is not necessary to reanalyze the base sequence.

(1 3 ) The gene expression analyzing means in the present invention preferably includes means for adding related information relating to the peak that has not been matched by the peak matching means. 4 ) The related information is at least one of an evaluation value, a peak position, a primer set, an expression intensity, a peak shape characteristic, and sample cell information and experimental information based on a correlation coefficient related to the similarity of the waveform. Is preferably included.

  If it does in this way, since it can add the related information regarding the peak about the peak which was not able to be matched in the above-mentioned peak matching process by the gene expression analysis means, for example, when referring to this later It is possible to obtain evaluation values based on correlation coefficients related to waveform similarity, peak positions, primer sets, expression intensity, peak shape characteristics, and related information such as sample cell information and experimental information. Therefore, the analysis result obtained in this way has a higher utility value for the user.

( 15 ) A gene expression analysis program according to the present invention is characterized by causing a computer to execute the gene expression analysis method described in (1) to ( 7 ).

As described above, since the gene expression analysis method described in (1) to ( 7 ) is executed by a computer, the reference profile data and the measurement profile data can be easily compared by the computer, and these peaks can be corresponded. In addition, the expression state of the gene can be analyzed.

  According to the gene expression analysis method of the present invention, the reference profile data and the measurement profile data are represented as waveforms and compared. Therefore, the comparison can be easily performed, and the gene expression state can be easily analyzed. Can do.

  Further, according to the gene expression analysis apparatus of the present invention, the reference profile data and the measurement profile data are represented as waveforms and compared, so that these comparisons can be easily performed and the gene expression state can be easily analyzed. It can be carried out.

  According to the gene expression analysis program of the present invention, the computer is executed so that the reference profile data and the measurement profile data are represented as waveforms and compared, and the peaks are associated with each other to analyze the gene expression state. Can be made.

  For example, measurement profile data obtained using a DNA sequencer based on the HiCEP method, which is a comprehensive and highly accurate gene expression analysis method, expressed as a waveform having a plurality of peaks, The abundance of mRNA derived from a specific gene (precisely, a relative value in one profile) is shown. By observing the intensity of all the peaks, it is possible to simultaneously measure highly accurate expression levels of tens of thousands of mRNAs. However, what kind of base sequence a peak has and what kind of gene it is from cannot be known unless the peak is sorted and the base sequence is decoded. It is important to know the gene (mRNA name) of the peak in order to connect a large amount of known gene information with the results of experiments using the HiCEP method. The gene information refers to the name of the transcript transcribed from the gene (may be a list if there are multiple genes), the function of the gene, and other attached information, including the transcript information described later. .

  The present invention expresses measurement profile data created by the HiCEP method or the like as a waveform, performs waveform correction processing based on a waveform whose genetic information is examined (that is, the waveform of reference profile data), and positions of the peaks , Or by adding or deleting one or more peaks from the measurement profile data to correct the waveform and aligning them, the peaks in the measurement profile data by HiCEP method correspond to the genetic information with high accuracy. In addition, the present invention embodies a gene expression analysis method that can identify a gene from which a peak is derived and analyze the expression state of the gene.

The best mode for carrying out a gene expression analysis method, a gene expression analysis apparatus, and a gene expression analysis program according to the present invention will be described below in detail with reference to the drawings as appropriate.
First, the gene expression analysis method according to the present invention will be described with reference to FIG. FIG. 1 is a flowchart for explaining the steps of the gene expression analysis method according to the present invention.

The gene expression analysis method according to the present invention is a gene expression analysis method for analyzing a gene expression state using profile data represented as one waveform having a plurality of peaks derived from a plurality of expressed gene transcripts. is there.
As shown in FIG. 1, the gene expression analysis method according to the present invention includes a reference profile data acquisition step S1, a measurement profile data creation step S2, a peak association step S3, and a gene expression analysis step S4. .

Here, “expression” in this technical field generally means that mRNA is transcribed from genomic DNA and translated as a protein, but “expression” in the present invention means gene transcription from genomic DNA. The product, typically mRNA, is transcribed.
In the present invention, “profile” refers to information including the gene expression pattern of a sample under a certain condition, the presence or absence of expression of a known and unknown gene, the expression level of all the expressed genes, and the like. “Profile data” refers to data representing such information as a single waveform having a plurality of peaks. Further, the “peak” means a part of the waveform described above, which is expressed, for example, in a substantially triangular shape based on the number of bases in the base sequence of the DNA fragment and the detected amount.

  As shown in FIG. 1, the reference profile data acquisition step S1 is a value corresponding to the number of bases of a DNA fragment such as a PCR amplification product obtained by amplifying a part of cDNA reverse-transcribed from a plurality of gene transcription products such as mRNA. The reference range of the position and the base number equivalent value obtained based on the detected amount corresponding to the transcription amount of the gene transcript at that position is represented as the first waveform, and the transcript species information of the gene transcript is represented by the first This is a step of previously acquiring reference profile data in which a predetermined peak in a waveform and a transcription product species from which the peak is derived are identified and stored. Here, the transcript type information refers to information on the type of gene transcript, for example, the peak position and peak intensity (height) are measured, the base sequence is determined, and the sequence name and accession number are It is given and the name of the gene of origin is specified. Information such as the name of the reported publication, the name of the author, and the name of the institution may also be included.

  Note that the reference profile data used in the present invention is configured to include peaks having link information to a database storing gene information, but the transcript species derived from all peaks in the reference profile data ( And / or gene) can be used as reference profile data even if they have not been determined. Furthermore, a plurality of profile data may be used, or only the profile portion (peak portion) of interest may be extracted and used. Further, reference profile data may be created by combining a plurality of waveforms, or reference profile data may be created artificially using only the peak position.

  Such reference profile data is acquired from a database storing known profile data, or is artificially created from transcript type information, or one or more known profile data or measurement profile data. Can be obtained by adding or deleting the peaks, or by combining a plurality of reference profile data, or by combining a plurality of measurement profile data. That is, it is only necessary to obtain profile data that is to be associated with measurement profile data in the subsequent peak association step S3 shown in FIG. Therefore, the present invention is not limited to using profile data prepared in advance. As described above, one of a plurality of created measurement profile data is used as reference profile data, and the other is used as measurement profile data. be able to.

  However, for genes derived from several peaks in the known profile data stored in the database, the base sequence has already been decoded, and the gene information for that peak, that is, the transcript type information of the gene transcript is not available. If added, these pieces of information can be obtained simultaneously by referring to known profile data. Therefore, referring to the known profile data is more advantageous as reference profile data in that one or more peaks are added to or deleted from the measurement profile data in that the transcript type information can be obtained. Is also suitable.

  The reference profile data used preferably refers to data obtained from the same species as a specific cell (sample) under certain conditions. The closer the conditions and state of the sample such as the strain and the derived tissue (cell) are, the higher the similarity between the measurement profile data (measurement profile data will be described in the measurement profile data creation step S2 described later) and the reference profile data. This is more effective. However, in the present invention, it is sufficient even when the measurement profile data of the sample and the reference profile data of the cell to be referenced are relatively close to each other, such as using the waveform of a similar cell as the reference profile data. It is effective (see FIG. 2).

2A to 2H show the HiCEP method for different cell lines A and B derived from the same species (specific description of the HiCEP method will be described later in measurement profile data creation step S2). FIG. 6 is a diagram comparing measurement profile data according to (1), wherein a combination of N ₁ N ₂ of the X adapter and N ₃ N ₄ of the Y adapter, (a) is for cell A (AA_AA), ( b) for cell B (AA_AA), (c) for cell A (CT_TC), (d) for cell B (CT_TC), (e) for cell A (AG_CG), (f) for cell B (AG_CG) and (g) are performed for cell A as (AG_CC), and (h) is performed for cell B as (AG_CC).
As shown in FIGS. 2A to 2H, it can be seen that the measurement profile data created by the HiCEP method has a similar waveform even in different cell lines.

  As shown in FIG. 1, in the next measurement profile data creation step S2, a PCR amplification product or the like that becomes a measurement object obtained by amplifying a part of cDNA obtained by reverse-transcription of a plurality of gene transcription products such as mRNA is obtained. Create measurement profile data that shows the measurement range of the base number equivalent value obtained based on the position corresponding to the base number of the DNA fragment and the detected amount equivalent to the transcription amount of the measurement object at that position as a second waveform It is a process to do.

The measurement profile data can be more suitably represented as a waveform by performing extrapolation complementation, exclusion from the analysis target, designation and exclusion of noise levels, separation of overlapping peaks, and the like.
Extrapolation complementation is because the expression level is too large and the DNA sequencer's sensor is saturated, and the waveform may become trapezoidal, or the peak may become a huge peak in the center. This means that the original peak is estimated in advance and the shape of the peak is complemented and expressed as a complete peak.

  In addition, the exclusion from the object of analysis means that in the HiCEP method using a capillary DNA sequencer, each color of ATGC is generated by simple scattered light of laser light that excites fluorescence when impurities or dust such as some crystals are precipitated are mixed. Are simultaneously measured, and a peak is detected at the same position in a plurality of profiles with the same intensity. Therefore, such an abnormal peak is excluded from the analysis target. Note that such an abnormal peak can also be detected from the shape of the peak such as how the peak spreads (σ) and the target property of the peak.

  The designation and exclusion of the noise level means that a small peak that should be regarded as noise is designated and excluded. That is, since the gene expression level is very small, a large number of small peaks are obtained, and therefore, such small peaks are excluded from the analysis target as noise. For example, when the number of gene transcripts is 5 / cell or less, it can be excluded from the analysis target.

  Overlapping peak separation means that a small peak is buried in the base of a large peak, or another peak appears in the shoulder of the peak, so the overlapping peak is a waveform compared to a single peak. This means that such a peak is separated because it is distorted. Such overlapping peaks can be predicted from waveform distortion by function approximation or waveform analysis. For example, they can be stored in a database or the like as one piece of information that the peak has. Accordingly, it can be added as related information regarding the peak in the gene expression analysis step S4 described later.

If there are overlapping peaks in the measurement profile data, the individual peaks can be separated and represented by the following.
(1) For the portion where the peaks overlap, the waveform may be created by adding the contributions of the overlapping peaks. This method is preferable because, in principle, it matches the waveform actually measured with high accuracy.
(2) It may be created by ignoring the overlap of peaks and switching to the function form of the counterpart at the intersection of the overlapping peak shapes. According to this method, peaks can be separated quickly and the peaks can be easily seen, but the shape may be different from the actual measurement waveform.

  In addition, when multiple peaks completely overlap even with these methods, only the largest peak can be seen. In such a case, the peak is separated and analyzed by decoding the base sequence, etc. It is good to do.

  Here, the position corresponding to the number of bases of the DNA fragment to be obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts (for example, mRNA), and the position of the measurement object at that position The measurement profile data representing the measurement range of the base number equivalent value obtained based on the detection amount equivalent to the transcription amount as the second waveform can be obtained by, for example, as shown in FIG. FIG. 3 is an explanatory diagram showing an example of a procedure for obtaining measurement profile data.

  As shown in FIG. 3A, first, a single strand cDNA (First strand cDNA) 13 in which a tag substance 12 such as biotin is added to the 5 ′ end using poly (A) RNA (mRNA) 11 as a template. Is synthesized (step a). Next, as shown in FIG. 5B, a double-stranded cDNA 14 is obtained using the single-stranded cDNA 13 synthesized in the step a as a template (step b). Then, as shown in FIG. 4C, the double-stranded cDNA 14 obtained in the step b is cleaved with a first restriction enzyme X (for example, a restriction enzyme with 4-base recognition) (step c). As shown in d), using a substance having a high affinity for the tag substance, the cDNA fragment to which the tag substance is added is recovered from the cDNA fragment obtained in the c process (d process).

  Next, as shown in FIG. 5 (e), a sequence complementary to the sequence of the cleavage site and a sequence complementary to the X primer to the cleavage site of the cDNA fragment recovered in step d by the first restriction enzyme X To obtain a cDNA fragment in which the X adapter 16 containing is bound to the 5 ′ end (step e), and the cDNA fragment obtained in step e does not cleave the X adapter 16 as shown in FIG. Cleavage with a second restriction enzyme Y (for example, a restriction enzyme with 4-base recognition) (step f). Then, as shown in FIG. 5G, using the substance 15 having high affinity for the tag substance 12, the cDNA fragment to which the tag substance 12 is bound is removed from the cDNA fragment obtained in step f. Thus, the cDNA fragment containing the cleavage site by the second restriction enzyme Y is recovered (step g), and the cDNA fragment recovered in the step g is recovered by the second restriction enzyme Y as shown in FIG. A cDNA fragment 18 is obtained in which a Y adapter 17 containing a sequence complementary to the sequence of the cleavage site and a sequence complementary to the Y primer is bound to the cleavage site (step h).

Next, as shown in FIG. 5 (i), N ₁ N ₂ (N ₁ and N ₂ are the same or different) containing a sequence complementary to the sequence of the X adapter 16 and having a 2 base sequence at its 3 ′ end. Which is a base selected from the group consisting of adenine, thymine, guanine and cytosine), and a sequence complementary to the sequence of the X adapter 19 containing the fluorescent substance 20 at the 5 ′ end and the Y adapter 17 N ₃ N ₄ (N ₃ and N ₄ may be the same or different, and is a base selected from the group consisting of adenine, thymine, guanine and cytosine) at the 3 ′ end thereof A PCR reaction using the cDNA fragment 18 obtained in step h as a template is performed using a primer set consisting of Y primer 21 (step i). Then, as shown in FIG. 6 (j), the PCR amplification product obtained in step i is applied to a DNA sequencer to detect the movement distance (that is, the number of bases) and the fluorescence amount (that is, the detection amount) of the PCR amplification product. (Step j) Measurement profile data can be created by expressing the detection result as a waveform.

In this way, the expression pattern of poly (A) RNA (mRNA) for a specific cell under a certain condition (whether it is known or unknown) is expressed as N ₁ , N ₂ , N _3. , N ₄ of four bases, that is, 4 ⁴ = 256 classifications (subgroups). However, the present invention is not limited to this. For example, N ₁ and N ₂ are N _A , N _B and N _C, and N ₃ and N ₄ are N _D , N _E and N _F 6 It may be a combination of two bases. In this case, it is possible to classify (subgroup) 4 ⁶ = 4096. Of course, the restriction enzymes X and Y used in this case must be 6-base recognition restriction enzymes.
Such a method for creating measurement profile data is described in, for example, pamphlet of International Publication No. WO2002 / 048352 and JP-A-2005-6554.

As shown in FIG. 1, in the next peak association step S <b> 3 to be performed next, a partial or entire region of at least one of the first waveform and the second waveform is corrected to adjust the peak position. By performing correction processing and associating the plurality of peaks including the region of interest in the second waveform described above with the plurality of peaks in the first waveform, the plurality of peaks including the region of interest and the plurality of peaks This is a step of associating a plurality of peaks in the first waveform corresponding to the peaks.
That is, for each peak included in these data, correction processing is performed to correct a part or all of one or both of the data and adjust the position of the peak, and peaks that are not at the same position as the peak at the same position are performed. Detect and associate peaks at the same position.

The peak associating step S3 is created in, for example, (1) the peak position of the reference profile data acquired in the reference profile data acquiring step S1 (see FIG. 1) and the measurement profile data creating step S2 (see FIG. 1). If the peak positions of the measured profile data match, the corresponding peak of the reference profile data is associated with the peak of the measurement profile data, and (2) the peak of the reference profile data acquired in the reference profile data acquisition step S1 When the position and the position of the peak of the measurement profile data created in the measurement profile data creation step S2 are partly or completely shifted, at least one of these waveforms has the highest similarity between these waveforms. Correct some or all of the area so that After performing the correction processing for integer, it can be suitably carried out by associating the peak of the reference profile data, the measured profile data and a peak.
Note that when the peak position of the measurement profile data is partially shifted, some of the multiple measurement profile data is shifted relative to the peak position of the reference profile data, This corresponds to a case where a certain size region (for example, between 200 bp and 350 bp) is shifted.

  As correction processing in the peak association step S3, a calculation reference point is prepared in advance, and the left and right sides of another reference point between the reference points are enlarged or reduced in the migration size (base number equivalent value) direction so that the waveforms are mutually connected. There is a global correction that corrects so that the evaluation value (similar to the correlation coefficient) is improved, and when the waveform peak is slightly shifted, the evaluation values before and after the peak (correlation coefficient) There is a local correction in which individual correction amounts are calculated and correction processing is performed so as to maximize the similarity). The former global correction can absorb relatively large measurement fluctuations derived from size marker recognition deviations and experimental conditions, while the latter local correction is derived from the three-dimensional structure of cDNA fragments during electrophoresis, etc. And relatively small measurement fluctuations derived from differences in experimental conditions.

  Therefore, as described above, one of the global correction and the local correction, or a correction process appropriately combining both, is performed according to the waveform state of the measurement profile data and the reference profile data. It can be adjusted so that the similarity of the waveform is the highest.

As an example of such correction processing, for example, (a) the peak position of the measurement profile data is moved with reference to the peak position of the reference profile data, or (b) the peak position of certain measurement profile data is used as a reference. Or (c) moving both the peak position of the reference profile data and the peak position of the measurement profile data. As a result, the similarity between the waveform of the measurement profile data and the waveform of the reference profile data can be adjusted to the highest level. As a result, the peak of the reference profile data and the peak of the measurement profile data can be handled with high accuracy. It becomes possible to attach.
The correction process described above can be corrected with higher accuracy as the number of waveforms to be corrected increases. In other words, you can move in a more plausible direction by majority vote. Therefore, when the number of waveforms to be corrected is sufficiently large, the position of the peak serving as the calculation reference point may be designated, or such a peak position may not be designated.

  Such correction processing can be performed simply and with high accuracy, and is preferably performed by function approximation based on a Gaussian function (Gaussian function approximation method). For example, the approximation is based on the Gaussian function approximation method. A trial subtraction method that repeats function approximation by sequentially subtracting the waveform contribution by the original data may be used together. In this way, more appropriate correction processing can be performed.

The correction process using both the Gaussian function approximation method and the trial subtraction method can be performed as follows, for example.
From the measured profile data expressed as a waveform, subtract the contribution of the main peak (the one whose approximation is likely to be certain in the first approximation) from the entire waveform, and apply the same waveform to the rest Approximate. Thereafter, this process is repeated until it falls within a predetermined range or exceeds a predetermined number of times. Such correction processing can be approximated with higher accuracy by performing approximation using the rising portions on both sides of the peak instead of using the base region. Further, in this case, it is preferable that the degree of the measurement point overlapping the waveform before the correction process in the corrected waveform is used as an evaluation criterion for the accuracy of the correction.

  When such correction processing is performed, first, correction processing using the Gaussian function approximation method is performed to list only certain peaks, and the results are displayed and automatically compared with criteria based on the user's rule of thumb. If it is determined that a higher-order approximate peak is necessary, the correction process is performed again using a trial subtraction method so that a peak with a lower evaluation value is acquired. can do.

  When there is a saturation peak (saturation), the following correction process may be performed. Since the saturation peak is a peak that is detected as a shape in which the tip is crushed due to the saturation state of the sensor of the measuring instrument, for example, the `` rising part '' and `` falling part '' that are the root parts of the saturation peak From this, a series of processes for estimating the shape of the tip of the central portion of the waveform and generating a peak of height that would originally exist using a Gaussian function or the like can be mentioned. Whether or not to perform such correction processing can be determined based on whether or not a threshold value set in consideration of the dynamic range of the device to be used is exceeded.

  The correction processing described above is performed on the waveform of the reference profile data and the waveform of the measurement profile data, and the positions of some or all of the peaks are adjusted so that the degree of similarity between these waveforms is the highest. An example in which the peaks of the reference profile data are associated will be described with reference to FIGS. 4 to 6 are diagrams showing measurement profile data (total 24 waveforms) created as a result of performing HiCEP method twice for 12 samples.

4A and 4B are examples of measurement profile data in which the positions of the peaks of the 24 waveforms are locally shifted. FIG. 4A shows a state before correction processing and FIG. 4B shows a state after correction processing.
In FIG. 4A, the two peaks at the center are shifted, and it is difficult to specify both peaks only by the peak position (base number equivalent value). However, the measurement profile data shown at the bottom of the waveform shown in the lower part of FIG. 5B is used as reference profile data, and the peak positions of the other 23 waveforms are used as a reference by the correction processing described above. When the adjustment is made, the center two peaks of the reference profile data and the center two peaks of the measurement profile data can be associated with each other with high accuracy as shown in the upper display section of FIG.

FIG. 5 is an example of measurement profile data in which the peak positions of 24 waveforms are largely shifted and shifted as a whole (FIG. 5 shows only a part of the measurement profile data). (A) is a figure before correction processing, (b) is a figure which shows the mode after correction processing.
In FIG. 5A, six of the 24 waveforms are greatly shifted in the direction of the peak position (value corresponding to the number of bases), and it appears that there are three peaks. However, as shown in FIG. 5B, one of the 24 waveforms is used as reference profile data, and the correction is performed as described above for the peak positions of the other 23 waveforms including the 6 waveforms described above on the basis of this. If it adjusts by processing, these will be gathered and it turns out that the position of a peak is two, and these can be matched accurately.

FIGS. 6A and 6B are examples of measurement profile data in which the positions of the peaks of 24 waveforms are close to each other. FIG. 6A shows a state before correction processing, and FIG. 6B shows a state after correction processing.
In FIG. 6A, since the peaks of 24 waveforms are close to each other and are combined, it is difficult to determine each peak just by slightly deviating the peak positions of these measurement profile data. However, the measurement profile data shown at the bottom of the waveform shown in the lower part of FIG. 5B is used as reference profile data, and the peak positions of the other 23 waveforms are used as a reference by the correction processing described above. When the adjustment is performed, all the peaks are aligned at a specific position as shown in the upper display section of FIG. 5B, and these can be associated with each other with high accuracy.

FIG. 7 is a diagram illustrating a case where a waveform having a peak at an arbitrary position is artificially created as reference profile data with reference to measurement profile data of 10 waveforms.
As shown in the upper display section of FIG. 7, with reference to measurement profile data of 10 waveforms, a waveform having a peak at an arbitrary position is artificially created as reference profile data, and 10 waveforms are based on this. If the position of the peak of the measurement profile data is adjusted by the correction process described above, all the peaks are aligned as shown in the lower display section of the figure, and these can be associated with high accuracy. As described above, the artificially created profile can be created based on the peak position information registered in the database. In this case, under different experimental conditions, the intensity value (peak height) can be arbitrarily set such as a peak with high expression, a small peak, and an intermediate peak. The peak created in this way can function effectively with respect to the correction process.

  Here, when gene information relating to the origin of the peak is added to the peak of the reference profile data, the gene described above is used for the peak of the measurement profile data that can be associated with the peak of the reference profile data. Information can be acquired from reference profile data and regarded as gene information of the peak of the measurement profile data.

As shown in FIG. 1, in the next gene expression analysis step S4, the information of the gene transcript derived from the peak of the associated measurement profile data is read from the transcript type information, and the associated measurement profile data It is a step of analyzing the expression state of a gene by specifying the gene from which the peak is derived.
In the gene expression analysis step S4, for example, the peak of the reference profile data and the peak of the measurement profile data are displayed so as to be distinguishable from the peak that cannot be matched, and the reference profile data If gene information about the gene from which the peak is derived is added to the peak, the gene expression state can be determined by identifying the gene of the peak associated in the measurement profile data by quoting the gene information. Can be analyzed. Note that displaying a peak that can be correlated with a peak that could not be correlated means, for example, displaying one peak in a different color or displaying one peak blinking. Can be mentioned.

  In the gene expression analysis step S4, if there is a peak that cannot be matched between the peak of the reference profile data and the peak of the measurement profile data, a more suitable analysis is performed by adding related information about the peak. Results are obtained. Such related information includes an evaluation value (see FIG. 8) based on a correlation coefficient regarding similarity to the waveform of the reference profile data, peak position, primer set, expression intensity, peak shape characteristics, and sample cells. Information and experimental information such as experimental conditions can be mentioned, and it is preferable that at least one of them is included. FIG. 8 illustrates an evaluation value based on a correlation coefficient related to similarity with the waveform of the reference profile data.

  As shown in the middle display section of FIG. 8, there may be a case where there is a peak not in the reference profile data in the measurement profile data, or there is no peak in the reference profile data in the measurement profile data. In this case, the evaluation value is low in relation to the reference profile data. Therefore, as shown in the vicinity of the center of the lower display portion in the figure, the evaluation value for the peak of the measurement profile data is slightly lower than the evaluation values for the other peaks. That is, the reliability regarding the peak size is slightly lower than the other peaks.

Of course, related information such as an evaluation value based on the correlation coefficient related to the similarity to the reference profile data may be added to the peak of the measurement profile data that can be associated with the peak of the reference profile data. .
As described above, when the evaluation value based on the correlation coefficient related to the similarity to the waveform of the reference profile data is added as the related information, the evaluation value is simultaneously added to the transcription product type information added to the measurement profile data. It is also possible to give an evaluation regarding the likelihood of the association.

  As described above, according to the gene expression analysis method of the present invention, since the reference profile data and the measurement profile data are represented as waveforms and compared, the comparison can be easily performed, and the expression state of the gene Can be easily analyzed. In addition, conventionally, the gene profile (mRNA) from which the peak of the measurement profile data is derived is determined each time the base sequence is determined and the similarity search is performed using an arbitrary database. If the gene information is added to the peak of the reference profile data used for matching, the gene from which the peak of the measurement profile data is derived can be identified. Therefore, the above-described work can be omitted.

  Next, the gene expression analysis apparatus according to the present invention will be described with reference to FIG. FIG. 9 is a block diagram showing an example of the configuration of the gene expression analysis apparatus according to the present invention.

A gene expression analysis apparatus A according to the present invention analyzes a gene expression state using profile data represented as a single waveform having a plurality of peaks derived from a plurality of expressed gene transcripts. In addition, as shown in FIG. 9, the apparatus includes reference profile data acquisition means 1, measurement profile data creation means 2, peak association means 3, and gene expression analysis means 4.
Note that the reference profile data acquisition means 1, measurement profile data creation means 2, peak association means 3 and gene expression analysis means 4 of the gene expression analysis apparatus A according to the present invention are respectively shown in FIG. Since this corresponds to the reference profile data acquisition step S1, the measurement profile data creation step S2, the peak association step S3, and the gene expression analysis step S4 of the expression analysis method, in the following description, the same contents as those described above The description about is omitted.

  The gene expression analyzer A can use, for example, a general-purpose computer or a workstation that is generally used. A DNA sequencer DS such as an external capillary DNA sequencer or a communication such as the Internet by a connection means (not shown). It is connected to the reference profile database DB1, the gene information database DB2, and the like via the network NW and can acquire necessary information in a timely manner. In addition, the gene expression analysis apparatus A is provided with a storage means 5 such as a hard disk drive, and stores or reads out the reference data, created data, analysis results, etc., as required by each means as described above. can do.

  The reference profile data acquisition unit 1 acquires in advance reference profile data that is a reference target of the measurement profile data created by the measurement profile data creation unit 2 to be described later from the reference profile database DB1 via the communication network NW. As described above, the reference profile data is the position corresponding to the number of bases of the DNA fragment obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts, and the amount corresponding to the transcription amount of the gene transcript at that position. The reference range of the base equivalent value obtained on the basis of the detected amount is expressed as the first waveform, and as the transcript type information of the gene transcript, the predetermined peak in the first waveform and its peak are derived It is the data which identified and memorize | stored the transcription | transfer product kind to perform. In addition, when gene information such as a gene name is added to the peak included in the reference profile data, or when it is set to be added, the gene information is transferred to the gene via the communication network NW. You may acquire from information database DB2.

  Here, the reference profile database DB1 and the gene information database DB2 are not limited to unique databases, such as a so-called database pool in which data that can be used as reference profile data is stored in a plurality of different databases. Also included. In such a case, in order to acquire the reference profile data, necessary data is appropriately acquired from a plurality of databases. Note that the user can arbitrarily determine what data is used as reference profile data. For example, if there are many peaks, the closest biological material, or the experimental condition, and the one with the most genetic information associated with it, it is obtained from multiple databases as reference profile data. be able to. For example, when the measurement profile data is a small number such as one or two, if only one reference profile data is acquired, the accuracy of waveform correction processing may not be good. In such a case, it is preferable to extract a plurality of relatively close data as reference profile data as described above, perform waveform correction processing, and use it for peak matching.

  As reference profile database DB1, for example, HiCEP database (URL: http://133.63.22.11/peakdb/query?request=dbmain&lang=ja) provided by National Institute of Radiological Sciences (Japan) Can be mentioned. As the gene information database DB2, for example, GenBank (provided by NCBI (National Center for Biotechnology Information) (USA) URL: http://www.ncbi.nlm.nih.gov/) URL: http://www.ncbi.nlm.nih.gov/Genbank/index.html), dbSNP (URL: http: //www.ncbi.nlm.nih. Database) of single nucleotide polymorphism (SNP) gov / projects / SNP /), EST (Expressed Sequence Tag) database dbEST (URL: http://www.ncbi.nlm.nih.gov/dbEST/), literature database MEDLINE (PubMed (URL: http://www.ncbi.nlm.nih.gov/sites/entrez?db=PubMed)).

  The measurement profile data creation means 2 is obtained by, for example, amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcription products to be measured by PCR method or the like as described with reference to FIG. Sequence result SR including the position corresponding to the number of bases measured by applying the DNA fragment (PCR amplification product) to be measured to the DNA sequencer DS and the detected amount corresponding to the transcription amount of the measurement object at that position For example, the measurement range of the base number equivalent value obtained based on the sequence result SR is, for example, a second peak having a plurality of peaks in the form of a graph with the horizontal axis as the base number and the vertical axis as the detection amount. Create measurement profile data expressed as a waveform.

  The peak associating means 3 performs a correction process for correcting a part or all of at least one of the first waveform and the second waveform, and adjusting the peak position, thereby obtaining the second waveform. By associating the plurality of peaks including the region of interest in the first peak with the plurality of peaks in the first waveform, the plurality of peaks including the region of interest and the first waveform corresponding to the plurality of peaks in the first waveform Associate multiple peaks.

  The gene expression analysis means 4 reads the information of the gene transcription product from which the peak of the associated measurement profile data is derived from the transcript type information, and identifies the gene from which the peak of the associated measurement profile data is derived. Analyze gene expression status. In addition, when adding gene information to the peak of the measurement profile data matched by the peak matching means 3, the gene expression analysis means 4 accesses the gene information database DB2 and stores the gene information related to the peak. You may make it add.

  Then, by outputting the analysis result analyzed as described above to the display means 6 such as a display or a printer, the expression state of the gene can be presented to the user.

  The gene expression analysis program according to the present invention is recorded on a computer-readable recording medium (not shown) such as a CD-ROM or a flexible disk, and is, for example, a recording medium driving apparatus connected to the gene expression analyzing apparatus A (Not shown), by reading the gene expression analysis program from the recording medium and installing it in the storage means 5, the reference profile data acquisition step S 1, measurement profile data creation step S 2 shown in FIG. The attaching step S3 and the gene expression analyzing step S4 may be executed.

  When the gene expression analysis program is stored in another computer (server) connected via the communication network NW, the gene expression analysis apparatus (client) connected to the communication network NW is By downloading the gene expression analysis program via the communication network NW, the computer may execute the reference profile data acquisition step S1, the measurement profile data creation step S2, the peak association step S3, and the gene expression analysis step S4. Good.

  The gene expression analysis method, the gene expression analysis apparatus, and the gene expression analysis program of the present invention have been described above in detail according to the best mode for carrying out the invention. However, the gist of the present invention is not limited to this. Needless to say, it should be interpreted widely based on the description of the scope of claims.

It is a flowchart explaining the procedure of the process of the gene expression analysis method which concerns on this invention. (A)-(h) is the figure which compared the measurement profile data by HiCEP method about different cell line A and cell line B. FIG. It is explanatory drawing which showed an example of the procedure until obtaining measurement profile data. It is an example of the measurement profile data from which the position of the peak of 24 waveforms has shifted locally, (a) is a figure before correction processing, and (b) is a figure showing signs after correction processing. FIG. 4A is an example of measurement profile data in which the positions of the peaks of 24 waveforms are greatly shifted and shifted as a whole, (a) showing a state before correction processing and (b) showing a state after correction processing. It is an example of the measurement profile data which the position of the peak of 24 waveforms adjoined, Comprising: (a) is a figure which shows the mode before correction processing, (b) is after correction processing. It is a figure which shows the case where the waveform which has a peak in arbitrary positions is created artificially and it is set as reference profile data with reference to the measurement profile data of 10 waveforms. The evaluation value based on the correlation coefficient regarding the similarity with the waveform of reference profile data is illustrated. It is a block diagram which shows an example of a structure of the gene expression analyzer which concerns on this invention.

Explanation of symbols

S1 Reference profile data acquisition step S2 Measurement profile data creation step S3 Peak association step S4 Gene expression analysis step A Gene expression analyzer 1 Reference profile data acquisition 2 Measurement profile data creation 3 Peak association step 4 Gene expression analysis step 5 Storage means 6 Display means DS DNA sequencer SR Sequence results (number of bases and detection amount of DNA fragment (PCR amplification product))
NW communication network DB1 reference profile database DB2 gene information database

Claims (15)

A gene expression analysis method for analyzing a gene expression state using profile data representing a plurality of peaks derived from a plurality of expressed gene transcripts as one waveform,
Obtained based on the position corresponding to the number of bases of a DNA fragment obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts and the detected amount corresponding to the transcription amount of the gene transcript at that position. The reference range of the base number equivalent value is represented as a first waveform, and as a transcription product species information of the gene transcript, a predetermined peak in the first waveform, a transcription product species from which the peak is derived, A reference profile data acquisition step of acquiring reference profile data that has been identified and stored in advance;
A position corresponding to the number of bases of a DNA fragment as a measurement object obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts, and a detection amount corresponding to the transcription amount of the measurement object at that position; The measurement profile data creating step for creating measurement profile data representing the measurement range of the base equivalent value obtained based on the second waveform,
A plurality of regions including a region of interest in the second waveform by performing correction processing for correcting a part or all of at least one of the first waveform and the second waveform and adjusting a peak position; And a plurality of peaks in the first waveform corresponding to the plurality of peaks, and a plurality of peaks in the first waveform corresponding to the plurality of peaks. A peak matching step to associate,
The gene expression state by reading the information of the gene transcript from which the peak of the associated measurement profile data is derived from the transcript type information and specifying the gene from which the peak of the associated measurement profile data is derived A gene expression analysis process for analyzing
Only including,
The peak association step includes
If the peak position of the reference profile data acquired in the reference profile data acquisition step matches the peak position of the measurement profile data created in the measurement profile data creation step, the matching reference Corresponding the peak of the profile data and the peak of the measurement profile data,
When the peak position of the reference profile data acquired in the reference profile data acquisition step and the peak position of the measurement profile data created in the measurement profile data creation step are partially or completely shifted Then, with respect to at least one of these waveforms, the reference profile data is corrected by correcting a part or all of the regions so that the similarity between these waveforms is the highest and adjusting the peak position. And a peak of the measurement profile data are associated with each other. The reference profile data acquisition step includes
The reference profile data is
Either from a database that stores known profile data,
Obtained by artificially creating from the transcript species information,
Acquired by adding or deleting one or more peaks in the known profile data or the measurement profile data,
Obtained by combining using a plurality of the reference profile data, or
The gene expression analysis method according to claim 1, wherein the gene expression analysis method is obtained by synthesizing using a plurality of the measurement profile data. The gene expression analysis method according to claim 1 or 2 , wherein the correction processing in the peak association step is performed by function approximation based on a Gaussian function. The correction process in the peak association step is
Moving the position of the peak of the measurement profile data based on the position of the peak of the reference profile data,
Moving the peak position of the reference profile data based on the peak position of the measurement profile data, or
Move both the peak position of the reference profile data and the peak position of the measurement profile data,
The gene expression analysis method according to any one of claims 1 to 3 , wherein a peak of the measurement profile data is associated with a peak of the reference profile data. The gene expression analysis step includes
The peak of the reference profile data and the peak of the measurement profile data are displayed so as to be distinguishable from the peak that cannot be correlated, and the peak of the reference profile data When gene information on the gene from which the gene is derived is added, the gene of the peak associated in the measurement profile data is identified by quoting the gene information, and the expression state of the gene is analyzed. The gene expression analysis method according to any one of claims 1 to 4 . In the gene expression analysis step,
The gene according to any one of claims 1 to 5 , further comprising a step of adding related information related to the peak for a peak that could not be matched in the peak matching step. Expression analysis method. The related information includes at least one of an evaluation value based on a correlation coefficient related to the similarity of the waveform, a peak position, a primer set, an expression intensity, a peak shape characteristic, and sample cell information and experimental information. The gene expression analysis method according to claim 6 , wherein: A gene expression analyzer that analyzes gene expression using profile data that represents a plurality of peaks derived from a plurality of expressed gene transcripts as a single waveform,
Obtained based on the position corresponding to the number of bases of a DNA fragment obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts and the detected amount corresponding to the transcription amount of the gene transcript at that position. The reference range of the base number equivalent value is represented as a first waveform, and as a transcription product species information of the gene transcript, a predetermined peak in the first waveform, a transcription product species from which the peak is derived, Reference profile data acquisition means for acquiring in advance the reference profile data that has been identified and stored;
A position corresponding to the number of bases of a DNA fragment as a measurement object obtained by amplifying a part of cDNA obtained by reverse transcription of a plurality of gene transcripts, and a detection amount corresponding to the transcription amount of the measurement object at that position; Measurement profile data creating means for creating measurement profile data representing the measurement range of the base number equivalent value obtained based on the second waveform,
A plurality of regions including a region of interest in the second waveform by performing correction processing for correcting a part or all of at least one of the first waveform and the second waveform and adjusting a peak position; And a plurality of peaks in the first waveform corresponding to the plurality of peaks, and a plurality of peaks in the first waveform corresponding to the plurality of peaks. A peak matching means to associate;
The gene expression state by reading the information of the gene transcript from which the peak of the associated measurement profile data is derived from the transcript type information and specifying the gene from which the peak of the associated measurement profile data is derived A gene expression analysis means for analyzing
I have a,
The peak association means includes
If the peak position of the reference profile data acquired by the reference profile data acquisition means matches the peak position of the measurement profile data created by the measurement profile data creation means, the matching reference Corresponding the peak of the profile data and the peak of the measurement profile data,
When the position of the peak of the reference profile data acquired by the reference profile data acquisition unit and the position of the peak of the measurement profile data generated by the measurement profile data generation unit are partially or completely shifted Then, with respect to at least one of these waveforms, the reference profile data is corrected by correcting a part or all of the regions so that the similarity between these waveforms is the highest and adjusting the peak position. And a peak of the measurement profile data are associated with each other. The reference profile data acquisition means includes
The reference profile data is
Either from a database that stores known profile data,
Obtained by artificially creating from the transcript species information,
Acquired by adding or deleting one or more peaks in the known profile data or the measurement profile data,
Obtained by combining using a plurality of the reference profile data, or
The gene expression analysis apparatus according to claim 8 , wherein the gene expression analysis apparatus is obtained by synthesizing using a plurality of the measurement profile data. The gene expression analysis apparatus according to claim 8 or 9 , wherein the correction processing in the peak association unit is performed by function approximation based on a Gaussian function. The correction process in the peak association means is
Moving the position of the peak of the measurement profile data based on the position of the peak of the reference profile data,
Moving the peak position of the reference profile data based on the peak position of the measurement profile data, or
Move both the peak position of the reference profile data and the peak position of the measurement profile data,
Gene expression analysis apparatus according to any one of claims 1 0 to claim 8, characterized in that associating the peak of the reference profile data and the peak of the measured profile data. The gene expression analysis means includes
The peak of the reference profile data and the peak of the measurement profile data are displayed so as to be distinguishable from the peak that cannot be correlated, and the peak of the reference profile data When gene information on the gene from which the gene is derived is added, the gene of the peak associated in the measurement profile data is identified by quoting the gene information, and the expression state of the gene is analyzed. The gene expression analysis apparatus according to any one of claims 8 to 11. The gene expression analysis means includes
Correspondence for the peak that could not at the peak correlating means, according to claims 8, characterized in that it contains means for adding the relevant information about the peak in any one of claims 1 2 Gene expression analyzer. The related information includes at least one of an evaluation value based on a correlation coefficient related to the similarity of the waveform, a peak position, a primer set, an expression intensity, a peak shape characteristic, and sample cell information and experimental information. gene expression analysis apparatus according to claim 1 3, characterized in that out. Gene expression analysis program characterized by executing the claims 1 to methods of gene expression analysis according to the computer to claim 7.

Images