JP5636614B2 - Comparative analysis method of data obtained by LC-MALDI - Google Patents

Description

  The present invention relates to a method for analyzing data acquired by matrix-assisted laser desorption / ionization mass spectrometry (hereinafter sometimes abbreviated as MALDI-MS) using liquid chromatography as a pretreatment.

  In recent years, with the completion of genome projects in various organisms and the development of mass spectrometers (hereinafter sometimes abbreviated as MS), proteome analysis that comprehensively analyzes organism proteins has been actively performed. There are various types of mass spectrometers depending on the combination of the ionization unit and the measurement unit, but the one that is widely used in the field of life science is the matrix-assisted laser desorption ionization (hereinafter sometimes abbreviated as MALDI) in the ionization unit. MALDI-MS adopting the method.

  In the MALDI method, a reagent called a matrix is mixed with an analysis sample, and a laser is applied to the sample to ionize the matrix molecules and the sample molecules. The matrix effectively absorbs the energy of the laser and assists ionization of the sample molecules. The MALDI method allows easy pretreatment of analytical samples, and allows analysis of various biomolecules by selecting a matrix. Therefore, mass spectrometers using this method are mainly used in biological and medical fields. Widely used.

  A notable breakthrough for the MALDI method is the improvement of Tanaka et al., Who won the Nobel Prize in Chemistry in 2002. The improvement of the matrix made by Tanaka et al. In 1987 enabled stable ionization of proteins that had previously been difficult to analyze, and promoted the spread of MALDI-MS in the life science field. The matrix at that time was a mixture of cobalt powder and glycerin, but now it is mainly organic compounds such as α-cyano-4-hydroxycinnamic acid (CHCA) and 2,5-dihydroxybenzoic acid (DHB). Is used.

  After the matrix is dissolved in the solvent, it is crystallized with the analytical sample on a conductive plate called a MALDI sample plate. The sample plate is inserted into a sample chamber in the mass spectrometer, where the mixed crystal is irradiated with laser. Matrix molecules and analysis sample molecules ionized by the energy of the laser are separated according to the mass and electric charge in the mass analysis unit, and reach the detector. The detected charges are processed in time series to become a mass spectrum.

  In order to perform mass spectrometry, it is necessary to purify an analytical sample in advance. Conventionally, separation by electrophoresis has been the main purification method, but in recent years, various chromatographic apparatuses are also used. An advantage of using a chromatographic apparatus for pretreatment is that a large amount of analysis sample can be separated with high resolution.

  Examples of an on-line system in which chromatography and a mass spectrometer are directly connected include LC / MS to which a liquid chromatography device (hereinafter referred to as LC) is connected, and GC / MS to which a gas chromatography device (hereinafter referred to as GC) is connected. Electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI) methods are used for the ionization part of LC / MS, and electron impact ionization (EI) and chemical ionization (CI) methods are used for the ionization part of GC / MS. Is done. In general, when referring to LC / MS or GC / MS, these are directly connected devices.

  LC / MS using the ESI method is excellent in throughput and completeness because it can sequentially ionize the liquid phase analysis sample separated by LC and send it to the mass spectrometer. Therefore, LC / MS is often used for large-scale proteome analysis. LC / MS has a large market scale, and freeware (Non-Patent Document 1) and commercial software compatible with various analysis methods are published (Non-Patent Documents 2 and 3).

  It is also possible to combine MALDI and chromatography. However, in the MALDI method, as described above, the analysis sample must be placed on the MALDI sample plate and crystallized once on the well together with the matrix. Is difficult. Therefore, an apparatus called a fraction collector is interposed between LC and MALDI-MS, and these are linked. The fraction collector is a device that separates a certain amount of sample eluted from the LC and automatically spots the wells on the MALDI sample plate in a certain order, and products are available from various companies (Non-Patent Document 4). , 5, 6, 7). A system in which LC and MALDI-MS are linked using such an apparatus is called LC-MALDI or LC-MALDI-MS.

  The advantage of LC-MALDI is that it is possible to perform proteomic analysis with high completeness using the popular MALDI-MS, and it is possible to separate LC and MALDI processes physically and temporally. In addition, it is possible to divide each process between distant laboratories and research institutions, to be able to repeatedly analyze the sample spotted on the MALDI sample plate, and to distribute ion to many wells to suppress the ionization characteristic peculiar to the MALDI method And that the MALDI method makes it difficult to generate multivalent ions, so that the interpretation and processing of the mass spectrum is easy.

  Conversely, the disadvantages of LC-MALDI are that a fraction collector is required, the throughput is inferior to LC / MS due to the process of spotting and crystallization, and the sample is spotted together at a fixed volume. The resolution of LC elution time is lost and many peaks are included in one spectrum. For the same reason, accurate elution time information from LC is lost for each peak.

  For the above reasons, LC-MALDI and general LC / MS differ in the nature and format of data obtained from the apparatus. Therefore, the above-mentioned data analysis software for LC / MS cannot be used for LC-MALDI. However, LC-MALDI is not widely used compared with LC / MS, and there is little data analysis software supported. Software for LC-MALDI is mainly used to control instruments and acquire data attached to fraction collectors and mass spectrometers (for example, the software package PROTEINEER-LC in Non-Patent Document 5). is there.

  Software specializing in LC-MALDI data analysis includes TWiP (Non-patent Document 8) manufactured by DYNACOM. TWiP is software for expression level fluctuation analysis using LC-MALDI and the NBS reagent kit (Non-patent Document 9), which is an isotope reagent of Shimadzu Corporation. A sample to be analyzed and a reference sample are labeled with isotopes having different mass numbers, and a homologous peak is detected based on a mass difference to capture a quantitative change.

  However, the method based on TWiP and isotope labeling recognizes only the peptide fragments present in both the sample to be analyzed and the reference sample in order to recognize the paired peak expected from the mass difference. There is a drawback that the presence of a specific peak that exists only in one sample is overlooked, which is the most significant difference between samples.

JP 2009-156722 A JP 2004-502934 A

“Metabolism control” 2005 research representative, Yoshiya Oda “Fusion of quantitative metabolomics and proteomics” 2008 results [online] URL http://www.jst.go.jp/kisoken/crest/report/heisei20 /pdf/pdf18/18-001.pdf LCMS // LCMS-IT-TOF Metabolomics analysis software Profiler [online] URL http://www.an.shimadzu.co.jp/lcms/it-tof6.htm Nonlinear Dynamics Progenesis LC-MS [online] URL http://www.scrum-net.co.jp/products/nonlinear/progenesis_lcms.htm [SHIMADZU BIOTECH] Spotting device for MALDI plate AccuSpot [online] URL http://www.shimadzu-biotech.jp/products/accu/index.html Product Information: PROTEINEER fc | Bruker Daltonics [online] URL http://www.bruker.jp/daltonics/product/proteineer-fc.html Yamato Science MALDI-TOF Direct Nano LC / MALDI Plate Spotting System for MS DiNa Map System [online] URL http://www.yamato-net.co.jp/support/catalog/pdf/p782_dina.pdf Agilent 1200 Series LC Microcollector / Spotter | Agilent Technologies, Inc. [online] URL http://www.chem-agilent.com/contents.php?id=38079 DYNACOM quantitative proteome analysis support software Twip [online] URL http://www.dynacom.co.jp/products/package/twip/twip_20060914.pdf [SHIMADZU BIOTECH] 13CNBS Stable Isotope Labeling Kit-N Stable isotope labeling kit for quantitative expression proteome analysis [online] URL http://www.shimadzu-biotech.jp/products/nbs/index.html

  An object of the present invention is to provide data analysis software using the system characteristics of LC-MALDI.

  In life sciences, it is often required to detect differences between two different samples. For example, it is an operation of finding a life phenomenon from the difference between the target group and the treatment group, such as a wild strain and a mutant strain, a normal state and a pathological state. The same applies to the situation where LC-MALDI is used such as proteome. The amount of data obtained by the proteome is enormous, and the use of computers is indispensable for detecting differences. The aforementioned TWiP is one of the software that answers such demands. However, homologous peak recognition using stable isotope labeling with TWiP only recognizes quantitative changes in peaks commonly found between two samples as described above, and detects a peak specific to one sample. There is a disadvantage that it can not.

  In addition, TWiP only supports analysis of the primary mass spectrum of peptide fragments obtained by digesting proteins with trypsin or the like, so-called precursor ion spectrum. Analysis of product ion spectra obtained by tandem mass spectrometry (hereinafter MS / MS) by post-source decomposition (PSD) or collision-induced dissociation (CID) is not expected. Rather, it is software that contributes to the step of selecting peaks that are candidates for detailed analysis, such as MS / MS.

  In order to solve the above problems, the present inventors have developed software for performing comparative analysis of a large amount of data obtained by combining LC-MALDI and MS / MS.

  In this specification, the term regarding mass spectrometry was based on the mass spectrometry related glossary (2009, ISBN 4-906661-02-2) edited by the Japan Society for Mass Spectrometry. An exception is the use of a mass-to-charge ratio not recommended in the glossary for m / z, which is a dimensionless quantity obtained by dividing the mass of an ion by the unit of atomic mass and the number of charges of the ion. There is. This is because there is no term that expresses m / z descriptively in Japanese, and in view of the circumstances that have been treated by those skilled in the art as having an equivalent mass-to-charge ratio with m / z. Therefore, in this specification, the mass-to-charge ratio and m / z are used interchangeably.

From the system combining LC-MALDI and MS / MS, various information about the analytical sample can be obtained. The following five are particularly important and related to the present invention.
1) Precursor ion m / z
2) The position on the sample plate where the precursor ion was obtained (specifically, the position of the well)
3) Product ion spectrum obtained by performing MS / MS on the precursor ion 4) m / z of the product ion group constituting the product ion spectrum
5) Peak intensity of the product ion group
A precursor spectrum is a primary spectrum composed of molecular groups obtained by irradiating a sample spotted in each well of a MALDI sample plate with a laser. Among the ions constituting this spectrum, an ion selected as a target of MS / MS is called a precursor ion. The precursor ion is uniquely identified by the m / z of the ion and the position of the well from which the ion was obtained.
A secondary spectrum obtained by performing MS / MS on a specific precursor ion is a product ion spectrum. Ions corresponding to individual peaks constituting the product ion spectrum are called product ions. In this specification, a group of product ions derived from one precursor ion is referred to as a product ion group. It is assumed that the data to be the object of the method of the present invention includes information of contents 1) to 5) or contents equivalent thereto.

  The present inventors considered that this information is useful as a parameter for judging the homology of data obtained from LC-MALDI, and paid attention to it. The feature of the present invention is that the position on the sample plate from which the precursor ion is obtained is used as one of the parameters for determining homology.

  The position on the sample plate where the precursor ions were obtained reflects the time when the fraction collector spotted the droplet containing the precursor ions on the well, that is, the time when the precursor ions were eluted from the LC. The elution time is a factor uniquely determined from the physical properties of the precursor ion and the implementation conditions such as the column and extraction solvent used in the LC. Therefore, when the execution conditions are the same, the same precursor ion is expected to be contained in the same or in the vicinity of the well, although there are disturbances due to differences in the analysis sample itself. Therefore, the positional information on the sample plate is useful as a parameter for judging the homology of the precursor ions contained therein. In JP-A-2009-156722 (Patent Document 1) and JP-A-2004-502934 (Patent Document 2), in LC-MALDI, the same component is present in several portions close to the LC retention time. It states that there is a possibility, but does not disclose any use or usefulness of the position information on the sample plate. In addition, since the location information on the sample plate is usually stored in the final MS / MS data, it is possible to improve the accuracy of homology determination using only the contents contained in the final MS / MS data. .

  In addition, after determining the homology by performing the second step of comparing product ion groups after the first step of performing the comparison based on the information of the precursor ions, the position information on the sample plate is used. By screening the precursor ions, the accuracy of the first step can be improved, the frequency of shifting to the comparison of the second step with a large amount of calculation can be reduced, and the time required for analysis can be shortened. In the comparison of the second step, brute force comparison between product ion groups is performed, so that the amount of calculation increases in proportion to the square of the number of product ions. By minimizing the number of times this is performed, it is possible to quickly analyze a large amount of data generated in the process of proteome analysis with less computer resources.

  Based on these ideas, the present inventors have established a method for determining the homology of data obtained from LC-MALDI. Furthermore, a method for collectively processing a large number of data has been established, and the present invention has been completed.

Therefore, according to one aspect of the present invention, the following method is provided.
The fractions of the sample separated by liquid chromatography are spotted on the sample plate in a predetermined order at predetermined positions, and these fractions are subjected to a tandem mass spectrometer equipped with an ionization part by matrix-assisted laser desorption ionization. (1) A precursor derived from a second sample relative to a position on a sample plate from which the precursor ion derived from the first sample is obtained. If the position on the sample plate from which the ions were acquired satisfies a predetermined relationship, and (2) the product ions of both precursor ions are compared, it is determined that both precursor ions are homologous. A method comprising the step of determining.

According to the method of the present invention, a pair of precursor ions to be compared (where one precursor ion is obtained from the first sample and the other precursor ion is obtained from the second sample) is the requirement of the above (1) And the above requirement (2) is satisfied, it is determined that there is homology. In this case, the determination of the requirement (1) and the determination of the requirement (2) may be preceded. However, as in the embodiment of the present invention described below, only when the requirement (1) is satisfied, the determination of the requirement (2) is more appropriate for the determination of the requirement (2) as described above. This is preferable because the calculation amount is reduced.
Thus, according to a preferred embodiment of the present invention, said step comprises
(1) The position on the sample plate from which the precursor ion derived from the second sample satisfies the predetermined relationship with respect to the position on the sample plate from which the precursor ion derived from the first sample is acquired. A first step of determining
(2) Second step of determining whether or not both precursor ions are homologous by comparing the product ions of both precursor ions determined to be within a predetermined range in the first step. When,
Comprising.

  In the present invention, it is required that the fraction of the sample separated by liquid chromatography is spotted on the sample plate in a predetermined order at a predetermined position, and which fraction is spotted on which position on the sample plate. It is only necessary to record the elution order of each spotted fraction, and it is not always necessary to spot adjacent fractions in adjacent wells of the sample plate. However, from the viewpoint of simplicity, it is preferable that the fractions are sequentially spotted in the adjacent wells of the sample plate in the order of elution, and the adjacent fractions are spotted in the adjacent wells. The first sample and the second sample are preferably separated by liquid chromatography under the same conditions. When the separation conditions of the first sample and the second sample are different, by adding a standard substance to each sample in advance or adopting a clearly assigned ion contained in the sample as an internal standard, It is necessary to detect a difference in elution time caused by a difference in separation conditions and to correct it by taking this into consideration for homology determination conditions.

  The first step may further include a step of determining that the mass-to-charge ratio of both precursor ions satisfies a predetermined relationship, as in the known method. That is, even if it is determined in the first step that the position on the sample plate from which both precursor ions are obtained satisfies the predetermined relationship, it is determined that the mass-to-charge ratio of both precursor ions does not satisfy the predetermined relationship. If it is determined that there is no homology, and if it is determined that the mass-to-charge ratio of both precursor ions satisfies a predetermined relationship, the homology may be determined in the second step.

  Comparison of the product ion groups of both precursor ions in the requirement (2) of the present invention can be usually performed by comparing the mass-to-charge ratio and the peak intensity of the product ion groups.

  The method of the present invention can be applied not only to one-to-one comparison of precursor ions but also to many-to-one and many-to-many comparisons. That is, when determining the homology between a plurality of precursor ions derived from the first sample and a single precursor ion derived from the second sample (many-to-one), After determining the homology between one precursor ion selected from the precursor ions and a single precursor ion derived from the second sample according to the method of the present invention, it is selected from a plurality of precursor ions derived from the first sample. The homology can be determined by repeating the procedure for determining the homology between the other precursor ion thus determined and the single precursor ion derived from the second sample according to the method of the present invention. Similarly, when determining homology between a single precursor ion derived from the first sample and a plurality of precursor ions derived from the second sample (many-to-one), or a plurality derived from the first sample In the case of determining the homology between a plurality of precursor ions derived from the second precursor ion and a plurality of precursor ions derived from the second sample (many-to-many), the homology can be determined by repeating the method of the present invention in the same manner as described above.

  Further, the present invention may further comprise a step of extracting a set of precursor ions determined to have homology or a set of precursor ions determined to have no homology from a plurality of precursor ions determined to be homologous. Furthermore, a display step of outputting the set to a display or a printed material may be provided.

The method of the present invention can be carried out using a computer. Therefore, according to another aspect, the present invention provides the following invention.
The fractions of the sample separated by liquid chromatography are spotted on the sample plate in a predetermined order at predetermined positions, and these fractions are subjected to a tandem mass spectrometer equipped with an ionization part by matrix-assisted laser desorption ionization. A computer program for determining the homology of the precursor ions obtained in the above, (1) with respect to the position on the sample plate from which the precursor ions derived from the first sample were obtained, When the position on the sample plate from which the precursor ion is derived satisfies the specified relationship, and (2) it is determined that both precursor ions are homologous by comparing the product ions of both precursor ions. A computer program comprising the step of determining that there is a sex, or Computer readable recording medium recording the computer program.

  According to the present invention, it is possible to compare data obtained by performing MS / MS using LC-MALDI and determine whether they are homologous or different with high accuracy. It became. In addition, it is possible to quickly compare a large amount of data and easily obtain a set of data having homology, or conversely, a set of data included only in a specific sample. By passing the data selected by the method of the present invention as a query to the database search program, it is possible to identify substances contained in a plurality of samples in common and substances contained only in a specific sample with high accuracy.

  In the present specification, the method and apparatus configuration of the present invention will be described below by taking an analysis of protein-derived peptides as an example. Since proteins are involved in many physiological processes in vivo and the databases to be searched are extensive, there are many opportunities to be analyzed, and this is a major target for the implementation of the method of the present invention. It is believed that there is. However, the application range of the present invention is not limited to protein-derived peptides. In view of the general usage of LC and MALDI-MS, other peptides, lipids, sugars, nucleic acids, proteins, synthetic polymers, and mixtures thereof, those modified, and multiple types of these bound As long as it is a sample that can be subjected to MS / MS in MALDI-MS after separation by LC, it can be a target for carrying out the method of the present invention.

  The format of data obtained by LC-MALDI is typically a text file in which ion information is described. In the Shimadzu equipment and software used in the examples of the present invention, in addition to the spectral file in the original format, the analysis results in the text file format corresponding to the database search software MASCOT and MASCOT Daemon (Matrix Science) can get. Some of them are shown in the examples. As a general-purpose data format whose specifications are publicly available, there is the mzData format (http://www.psidev.info/index.php?q=node/80#mzdata). However, the scope of application of the present invention is not limited to data recorded in a text file, and any format may be used as long as it includes information necessary for carrying out the method of the present invention. It is assumed that various types of data are handled depending on the type of mass spectrometer and software used for the analysis, and the user data management method. For example, the data may be a binary file or data stored as a record in a relational database instead of a file. When a plurality of data are handled, the data may be stored as a data list in a single file for each analysis sample, or may be a plurality of files divided in arbitrary units.

  When the method of the present invention is realized by software operating on a computer, the software realizing the present invention may be written in any programming language. For example, it can be selected according to the application of the practitioner, such as describing in an interpreter programming language when frequent code changes are required, and using a compiler programming language when processing a large amount of data at high speed.

  The hardware on which the software that implements the present invention operates may be any hardware as long as the software operates. Personal computers can be selected for small-scale analysis, and workstations and large computers can be selected for large-scale analysis, depending on the application of the practitioner. The hardware may be a general-purpose computer, or may be provided as a dedicated device incorporating software that implements the method of the present invention or an analysis system including the dedicated device.

  The “predetermined relationship” in the requirement (1) that needs to be set when carrying out the present invention (for example, the allowable limit of the difference in elution order between fractions from which a pair of precursor ions to be compared is derived) And a threshold value such as a “predetermined relationship” of the mass-to-charge ratio of the precursor ions determined as desired in the first step (for example, an allowable limit of the difference in mass-to-charge ratio of a pair of precursor ions to be compared) The practitioner can set as appropriate based on the accuracy of the analytical instrument to be used and the past experience of the practitioner. It is also possible to select an appropriate threshold value by repeating feedback while visually observing the result by referring to the result of implementation of the method of the present invention written in a comparison matrix diagram (specifically, described in FIG. 2 and the example). It is. The threshold value can also be set dynamically by automating the process of determining numerical values as described above by software and linking it with the method of the present invention.

  Details of the algorithm implementing the method of the invention are given in the examples. However, this exemplary algorithm is preferably extensible by several methods. In the embodiment, the first step and the second step are independent processes. For example, if there is a high probability that the precursor ions determined to satisfy the predetermined relationship in the first step are homologous. Lower the threshold used in the comparison in the second step, and conversely, if the probability that the precursor ions determined to satisfy the predetermined relationship in the first step are homologous is low, increase the threshold in the second step. Thus, modifications such as making a stricter judgment can be considered.

  The results compared by the method of the present invention can be output as a work log in a text format, or can be output as a visible comparison matrix diagram or a Venn diagram indicating the data inclusion relationship. In addition, a new term or field can be provided in the data file or record to be processed, and the determination result can be described there. Alternatively, if the performance of the entire process is emphasized, the process of outputting the comparison result can be omitted, and only a set of homologous data or a set of non-homologous data can be created.

The following experiment was conducted using the chloroplast division apparatus of the unicellular red alga Cyanidioschyzon merolae. With schizon, a method for isolating organelles has been established, culture that synchronizes the cell cycle with light conditions is possible, and genome information has been completely deciphered. Is a unicellular algae used as a model organism.
The analytical sample was prepared according to the method of Yoshida et al. 2006. Interphase cells and mitotic cells were sampled from a culture system of schizon whose cell cycle was synchronized by light-synchronized culture. After the cells were physically disrupted by a French press, the membrane system was lysed by surfactant treatment, and the chloroplast division apparatus was isolated by density gradient centrifugation. Solubilized chloroplast division apparatus derived from interphase cells and chloroplast division apparatus derived from mitotic cells obtained by the above steps with octyl glucoside, which is a nonionic surfactant, An analysis sample was used. These samples are hereinafter referred to as interphase samples and mitotic samples, respectively.

The equipment configuration of LC-MALDI is explained. LC was HPLC of Shimadzu Prominence series. LC-20Adnano was used for the liquid feeding unit, L-column ODS for the trap column, and monolith type ODS for the separation column. The detector used was a UV-visible detector MU701 from GL Sciences, Shimadzu AccuSpot as the fraction collector, and AXIMA-TOF2 from Shimadzu as the mass spectrometer.
Trifluoroacetic acid (TFA) and acetonitrile (MeCN) were used as the mobile phase of HPLC. The MeCN concentration was changed from 10% to 80% in 65 minutes, and the sample was eluted from the stationary phase at a flow rate of 1 μl / min. As the matrix, CHCA (solvent 0.1% TFA + 70% MeCN, flow rate at the time of addition 2.7 μl / min) was used. The eluted samples were sequentially spotted by AccuSpot every 10 seconds into wells on the MALDI sample plate, and 768 spots were obtained from interphase samples and mitotic samples, for a total of 768 spots. In addition, the order of spots followed the order of numbering of wells.

Mass spectrometry was performed in the automatic analysis mode of reflectron mode from Shimadzu Biotech Launchpad manufactured by Kratos Analytical, which is control software of AXIMA-TOF2. Analysis settings were performed as follows.
1) First, acquire a precursor ion spectrum for one well.
2) From the precursor ion spectrum, a maximum of 9 significant precursor ions are selected based on the peak intensity. If a precursor ion with a strength worthy of selection is not included, the following procedure is omitted, and the analysis proceeds to the next well.
3) MS / MS is performed on each precursor ion to obtain a product ion spectrum.
4) Select 40 product ions from the product ion spectrum based on the peak intensity.
5) Record the acquired information and move on to analysis of the next well.
6) Repeat the above procedure and analyze 384 wells.
Through this procedure, information such as m / z of the precursor ion, the position of the well from which the precursor ion was obtained, m / z of the 40 product ion groups constituting the product ion spectrum, and peak intensity were obtained.

  As a result of the analysis, 1770 precursor ions were selected for interphase samples and 2000 precursor ions for mitotic samples. Product ion spectra were obtained from these precursor ions, and m / z and peak intensity of 70800 product ions and 80000 product ions were obtained, respectively.

  The beginning of the data list obtained in this example is illustrated below. The file format is a text file.

#Generated by: Shimadzu Biotech Launchpad 2.9.0.20090423

BEGIN IONS
TITLE = "380-01-04-16-972-0001"
PEPMASS = 972.112698 #Monoisotopic parent mass converted to average mass
69.553451 208.782730
81.189221 227.407715
87.102240 194.800308
120.802460 199.514313
148.074630 208.867783
178.270482 215.040237
188.191324 200.349854
203.051050 243.305008
232.555640 400.067047
285.243322 220.309723
311.047473 239.137558
342.505211 227.958115
352.137630 221.912903
366.910394 279.571991
370.376326 221.070374
376.631525 231.490173
478.242245 385.577850
544.857414 374.178406
547.501031 438.071411
550.360622 377.747559
557.402083 388.775513
568.244924 416.930939
571.906657 459.858154
575.866285 418.673889
693.642359 727.524536
720.847259 736.569031
728.364812 775.131592
735.962377 925.092712
744.863465 928.978699
748.373898 916.462524
755.954325 880.141479
762.831294 718.646729
864.344841 1220.819458
882.299345 1636.807373
892.143029 1412.451660
909.402658 2141.953369
919.234859 1347.742920
926.522635 4387.279785
936.652930 1600.116455
945.636016 2003.659912
END IONS

BEGIN IONS
TITLE = "13-00-03-02-1985-0001"
PEPMASS = 1986.169291 #Monoisotopic parent mass converted to average mass
175.322745 459.046021
319.550525 285.396790
327.164492 293.305725
(Hereafter omitted)

  The structure of the data will be described. The first line starting with the # symbol is a comment line automatically added by the application. A line including a # symbol like this line is treated as a comment after the # symbol and is not processed. The BEGIN IONS line and the END IONS line indicate the start and end of ion information, and this interval is data for one precursor ion and the product ion group derived from it. The TITLE line is information including well position information and uniquely identifies a precursor ion. The PEPMASS line indicates the m / z of the precursor ion. The information from the line following the PEPMASS line to the line before END IONS is the product ion group information, the left number indicates the product ion m / z, and the right number indicates the product ion peak intensity.

  Software was created to implement the present invention. Perl (Active Perl v5.6.1), an interpreted programming language, was used for creation. The software was run on a personal computer equipped with Microsoft Windows® XP as the operating system.

  The flow of software processing is shown in FIG. 1 as a flowchart. In this example, the data group of the interphase samples was processed as data list A in the figure, and the data group of mitotic samples was processed as data list B. Data of interphase samples are extracted one by one from the beginning, and each of them is compared with all data of mitotic samples. The collation is performed from the first-stage comparison described later, and the second-stage comparison is performed only when the reference is exceeded.

Hereinafter, the process of data comparison will be described in detail. In the first stage comparison, the comparison of the precursor ion mass to charge ratio was based on Equation 1 below.
abs ($ divpepmass [$ i]-$ intpepmass [$ j]) <= $ pepmasstol Formula 1
Where $ divpepmass [$ i] is the m / z of the i th precursor ion of the mitotic sample and $ intpepmass [$ j] is the m / z of the j th precursor ion of the interphase sample. The absolute value of these differences was determined by the absolute value function abs and compared with the error tolerance $ pepmasstol. The value of $ pepmasstol was set to 0.3 based on the error tolerance at the time of data acquisition. In general, the value of $ pepmasstol is preferably set in the range of 0.1 to 0.5 in the reflectron mode.
Further, in the first stage comparison, the comparison of the well positions was based on the following Equation 2.
abs ($ divrt [$ i]-$ intrt [$ j]) <= $ rttolmin Equation 2
Where $ divrt [$ i] is the position of the well from which the i th precursor ion of the mitotic sample was obtained, and $ intrt [$ j] is the position of the well from which the j th precursor ion of the interphase sample was obtained It is. Similar to the mass-to-charge ratio, the absolute value of the difference was obtained and compared with the error tolerance $ rttolmin. The value of $ rttolmin was 38 based on 10% of the total number of wells (384). In general, the value of $ rttolmin is preferably set in the range of 3% to 15% of the total number of wells.
The above comparison was made in a round robin manner by scanning two-dimensionally from 1st to 1770th for i and 1st to 2000th for j. When both Formula 1 and Formula 2 were true, it was judged that the precursor ion similarity exceeded the standard, and a second-stage comparison was performed.

In the second stage comparison, the comparison of the mass-to-charge ratio of the product ions was based on Equation 3 below.
abs ($ divfragion [0] [$ i] [$ k]-$ intfragion [0] [$ j] [$ l]) <= $ fragmasstol expression 3
Where $ divfragion [0] [$ i] [$ k] is the m / z of the k th product ion derived from the i th precursor ion of the mitotic sample, $ intfragion [0] [$ j] [$ l] is the m / z of the l th product ion derived from the j th precursor ion of the interphase sample. The absolute value of these differences was determined and compared to the error tolerance $ fragmasstol. The value of $ fragmasstol was set to 1.3 based on the error tolerance at the time of data acquisition. In general, the value of $ fragmasstol is preferably set in the range of 0.5 to 2.0.
Further, in the second-stage comparison, the comparison of peak intensities was based on the following formula 4.
$ divfragion [1] [$ i] [$ k] + $ intfragion [1] [$ j] [$ l]> = $ pithreshold formula 4
Where $ divfragion [1] [$ i] [$ k] is the peak intensity of the kth product ion derived from the ith precursor ion of the mitotic sample, $ intfragion [1] [$ j] [$ l ] Is the peak intensity of the l-th product ion derived from the j-th precursor ion of the interphase sample. These sums were compared with the strength reference value $ pithreshold. $ pithreshold was set to 1500. Note that $ pithreshold is preferably set as appropriate according to the quality of the spectrum, the accuracy of peak recognition of product ions, a numerical algorithm, and the like.
The above comparison was performed in a round robin manner by scanning the first to the 40th of k and l two-dimensionally. When both Equation 3 and Equation 4 were true, it was determined that the product ion similarity exceeded the standard.

  The variable $ similarity was defined to record the results of the second comparison. $ similarity is a variable that is initialized to 0 each time when moving from the first-level comparison to the second-level comparison. If both Expression 3 and Expression 4 are true in the second-stage comparison, 1 is added to the value of $ similarity. When the comparison is completed for 40 product ions, if $ similarity is 10 or more, that is, if at least 10 product ions out of 40 are similar, the precursor ions are determined to be homologous.

As a result of the above treatment, 522 precursor ions were determined to be homologous between the interphase sample and the mitotic sample. That is, among 1770 interphase samples, 1248 were specific to interphase samples, and 1478 out of 2000 samples were also mitotic samples. The collation result was output as a comparison matrix and work log.
Based on this processing result, the data of the mitotic sample were divided into those specific to the mitotic sample and those that showed homology with the interphase sample. Thus, it is expected that precursor ions derived from proteins existing even in interphase are subtracted from the mitotic sample, and precursor ions derived from mitotic phase-specific proteins remain.

In order to confirm the effect of the treatment, using the data before the treatment of the mitotic sample (precursor ion number 2000) and the data obtained by subtracting the data that showed homology with the interphase sample (1478), schizon A MASCOT search was performed on the genome database. Table 1 shows the search results.
As a result of database search, the ranking of genes of known proteins involved in organelle division such as CMN262C and CME019C, and genes of GTP-binding proteins such as CMD101C and CMJ069C rose significantly after treatment. GTP-binding proteins are a family of proteins that bind to GTP and its hydrolyzate GDP and have various functions, and are involved in the regulation of biosynthesis and transport of cytoskeletal proteins and the like. Therefore, these GTP binding proteins are also expected to be related to the control of cell division.
Conversely, the genes of the photosynthetic proteins encoded by chloroplasts such as CMV157C, CMV051C, CMV063C, and CMV158C did not change or declined. This indicates that the precursor ions derived from these proteins are also present in the interphase sample, which has been subtracted by comparative analysis. Therefore, these proteins are expected to be non-specifically expressed proteins in the cell cycle.
Thus, the protein of the present invention was able to efficiently extract a protein that is expected to exist specifically in the mitotic phase.

Next, the superiority of adding well position information in determining the homology of precursor ions will be described with reference to the drawings. FIG. 2 illustrates a comparison matrix obtained as a result of collating the interphase sample and the mitotic sample with different settings. The horizontal axis represents the precursor ion of the interphase sample, and represents the 1st to 1770th in order from the left. The vertical axis represents the precursor ion of the mitotic sample, and represents the first to 2000th in order from the top. A black dot at the intersection of the vertical axis and the horizontal axis indicates that the corresponding precursor ions on the vertical axis and the horizontal axis are determined to be homologous. Detailed methods and parameters for comparison are in accordance with the above examples,
(I) is a matrix in the case where determination is made only by m / z of the precursor ion without considering the position of the well in the comparison of the precursor ion at the first stage. Black spots are concentrated in the sigmoid type from the upper left to the lower right, and it is expected that the black spots along this curve are determined to be homologous between the precursor ions that are truly homologous. However, in (I), many black spots are distributed throughout the matrix other than the vicinity of the curve. It is expected that these are due to false positive judgments that originally considered different precursor ions as homologous.
In order to reduce false positives, (II) shows the result of raising the threshold of judgment over the setting of (I) in the second stage product ion matching. Specifically, in the verification of 40 product ions, in (I), if 10 or more were determined to be homologous, the precursor ion was determined to be homologous, but this criterion was changed to 15 or more. is there. As a result, although the number of black spots scattered throughout the matrix is reduced, the black spots along the curve are also reduced. This is presumably due to an increase in false negative judgments that overlooked combinations that should be regarded as homologous. The reduction of black spots on the curve is particularly noticeable at the lower right of the matrix.
The threshold at the time of product ion matching is returned to 10, and the position of the well is taken into consideration in the first stage comparison (III). This setting is the same as in the above embodiment. In (III), compared with (I), the number of black spots scattered throughout the entire matrix was reduced, while the decrease in black spots along the curve was suppressed more than in (II). From these results, it was shown that adding well position information in the first-stage comparison is effective in determining the homology of precursor ions.

  The above embodiments are merely intended to assist the illustration and understanding of the present invention, and do not limit the scope of protection described in the claims.

(References for Examples)
Yoshida Y, Kuroiwa H, Misumi O, Nishida K, Yagisawa F, Fujiwara T, Nanamiya H, Kawamura F, Kuroiwa T.
Isolated chloroplast division machinery can actively constrict after stretching.
Science. 2006 Sep 8; 313 (5792): 1435-8.PMID: 16960006

The present invention is versatile as a method for processing data obtained by LC-MALDI, and can be used in various aspects. The present invention enables differential analysis of various samples such as biological tissue, extracts derived therefrom, or chemical industrial products. For example, by comparing the proteome of normal tissue and pathological tissue in the present invention, it is possible to find a factor that is specifically expressed in the latter or a factor that is lost in the latter, a so-called biomarker. . Incorporating this into the drug discovery process is expected to improve productivity in pharmaceuticals and discover new drug discovery targets. Also, in the chemical and food industries, by comparing multiple identical products, you can check product quality variations and contamination contained in the products, or compare new products with aged products over time. It is also possible to contribute to quality control, such as detecting composition changes due to
Conversely, if attention is paid to data common to a plurality of samples, factors common to various biological tissues can be found and used as controls and standard substances. In quality control, it is conceivable to use such methods as detecting universal contamination originating in the manufacturing environment by finding factors that are commonly included in different products.

The flowchart which shows the procedure of the analysis method of this invention in the case of comparing many-to-many data. The comparison matrix figure of a processing result which shows the usefulness of adding the positional information on a well in determination of the homology of a precursor ion.

Claims (11)

The fractions of the sample separated by liquid chromatography are spotted on the sample plate in a predetermined order at predetermined positions, and these fractions are subjected to a tandem mass spectrometer equipped with an ionization part by matrix-assisted laser desorption ionization. A method for determining the homology of the precursor ions obtained by
(1) The position on the sample plate from which the precursor ion derived from the second sample satisfies the predetermined relationship with respect to the position on the sample plate from which the precursor ion derived from the first sample is acquired. A first step of determining
(2) a second step of determining whether or not both precursor ions are homologous by comparing the product ions of both precursor ions determined to satisfy the predetermined relationship in the first step. ,
A method comprising the steps of: Said first step further, mass-to-charge ratio of both precursor ions includes the step of determining that a predetermined relationship, the method according to claim 1. The method according to claim 1 , wherein the comparison of the product ion groups of both precursor ions in the requirement (2) is performed by comparing the mass-to-charge ratio and the peak intensity of the product ion groups.   The method further comprises the step of taking out a set of precursor ions determined to have homology or a set of precursor ions determined to have no homology from a plurality of precursor ions determined to be homologous. the method of. 5. The method of claim 4 , further comprising displaying the set. The fractions of the sample separated by liquid chromatography are spotted on the sample plate in a predetermined order at predetermined positions, and these fractions are subjected to a tandem mass spectrometer equipped with an ionization part by matrix-assisted laser desorption ionization. A computer program for determining the homology of the precursor ions obtained by
(1) The position on the sample plate from which the precursor ion derived from the second sample satisfies the predetermined relationship with respect to the position on the sample plate from which the precursor ion derived from the first sample is acquired. A first step of determining
(2) a second step of determining whether or not both precursor ions are homologous by comparing the product ions of both precursor ions determined to satisfy the predetermined relationship in the first step. ,
Computer program characterized by causing a computer to execute the. It said first step further, mass-to-charge ratio of both precursor ions includes the step of determining that a predetermined relationship, the computer program of claim 6. (2) Comparison of product ions of both precursor ions in requirements is done by comparing the mass-to-charge ratio and the peak intensity of the product ion group, computer program of claim 6. Further, the homology of a plurality of precursor ions is determined, and a step of retrieving a set of homologous chromatic been judged precursor ion collection or homology No been judged precursor ion, according to claim 6 of computer programs. Furthermore, the computer program of claim 9 comprising comprising the step of displaying the set. 11. A computer-readable recording medium on which the computer program according to claim 6 is recorded.

Images