JP4541122B2 - Amino acid sequence identification method using mass spectrometry - Google Patents

Description

  The present invention relates to a novel method for identifying an amino acid sequence using a mass spectrum obtained by mass spectrometry.

  Mass spectrometry (mass spectrometry) uses molecular particles as gaseous ions, and these ions are moved in a vacuum electromagnetic field to separate and detect according to the mass-to-charge ratio (m / z) of the ions. Technology. An apparatus for performing the measurement is referred to as a mass spectrometer, and a detection signal obtained thereby is represented as a graph of ionic strength against mass-to-charge ratio.

  In protein identification analysis using mass spectrometry (see Non-Patent Document 1), peptide fragments digested by proteolytic enzymes are introduced into a mass spectrometer, and the mass-to-charge ratio of each peptide molecule ion is measured. Depending on the enzyme used, the position of digestion on the amino acid sequence of the protein is limited to a few specific locations. If the mass of the peptide molecule is determined from the MS spectrum of the peptide obtained as a protein digestion product, the amino acid sequence can be predicted by using the amino acid sequence database. This technique is called peptide mass fingerprinting. However, since it is difficult to determine the sequence uniquely by this method, the peptide sequence tag method using MS / MS spectrum is usually used.

  The MS / MS spectrum is a molecular ion (precursor ion) introduced into the mass spectrometer, which is randomly decomposed (cleaved) by collision-induced dissociation (CID), etc. It is obtained by mass spectrometry and contains information about the structure of the molecule before decomposition. It is known that a peptide in which a plurality of amino acids are connected in a straight chain is easily cleaved by CID at the three peptide binding positions by, ax, or cz shown in FIG.

 If this cleavage occurs uniformly on the peptide chain, all MS / MS spectra corresponding to the mass from the end to the a-x, b-y, c-z cleavage site will be obtained. Since the MS / MS spectrum from the terminal to the cleavage site shows the combined mass of each amino acid, the partial sequence of the peptide can be predicted. In fact, since cleavage does not occur at a uniform probability at all cleavable sites, in addition to the peptide mass fingerprint method, a method of searching an amino acid sequence database using partially read amino acid sequences Is the peptide sequence tag method.

  On the other hand, de novo estimates the corresponding amino acid sequence by reading as much as possible the a, b, c- or x, y, z-series signals present in the MS / MS spectrum without using any database There is also a sequencing method. That is, in the existing de novo sequencing method, spectra corresponding to individual amino acid residues are identified from the entire obtained MS / MS spectrum by combination calculation using graph theory and dynamic programming method, and the amino acid sequence of the peptide Is identified. However, the existing de novo sequencing method has a problem that it requires enormous calculation processing and cannot identify and identify amino acids having the same mass such as leucine and isoleucine.

Proteome analysis-advanced technology for protein expression and functional analysis and genomic medicine / drug discovery ISBN: 4897069335, Yodosha (published 2000-07-10) Page: 129-136

  Therefore, in view of the problems of the conventional amino acid sequence identification method as described above, an object of the present invention is to improve the efficiency of calculation processing and provide a highly accurate amino acid sequence identification method.

  Thus, as a result of intensive studies to solve the above-mentioned problems, the present inventors have determined that “unsuccessful” derived from the ionic strength of the MS / MS spectrum and the binding strength between amino acids, unlike the existing de novo sequencing method. It has been found that the amino acid sequence can be identified with high precision by using "", and the present invention has been completed.

  Further, the present inventor has found that the combination of amino acids constituting the precursor ion can be identified by using the storage means storing the mass of the precursor ion obtained from the MS / MS spectrum and the mass value for each amino acid. It came to be completed.

That is, the present invention includes the following.
(1) From an input means for inputting a mass spectrum consisting of a mass-to-charge ratio and ionic strength obtained from a sample containing a peptide to be sequence-identified, and a storage means for storing amino acid theoretical mass values for each amino acid. Calculate the estimated mass value based on the amino acid sequence of the peptide fragment up to d (an integer greater than or equal to 0) and add the theoretical mass value of the amino acid to the calculated estimated mass value. The estimated mass value of each peptide fragment (the number of amino acids up to an integer of 1 or more) is calculated, and a specific mass spectrum among the estimated mass values of the peptide fragments up to d + Nth and the mass spectrum input from the above input means Calculate the difference from the measured mass value calculated from the mass-to-charge ratio, identify the calculated difference within the specified range, and use the N amino acids used to calculate the identified difference With respect to each candidate amino acid searched by the candidate amino acid search means as candidate amino acids for the d + 1 to d + N-th amino acids, and the candidate amino acid searched by the candidate amino acid search means, the ion intensity of the specific mass spectrum is An evaluation value calculating means for calculating an evaluation value using an evaluation function that positively evaluates the case where the gap between the d-th amino acid and the candidate amino acid is easily cut off, and the candidate using the obtained evaluation value An amino acid sequence identification apparatus comprising: identification means for identifying amino acid sequences in peptide fragments up to d + N in the peptide by identifying one candidate amino acid from amino acids.

(2) It further comprises ion intensity probability value calculating means for scaling the ion intensity included in the mass spectrum input by the input means so as to be uniformly distributed within a random variable, and the evaluation value calculating means includes the ion intensity probability. The amino acid sequence identification device according to (1), wherein the evaluation value is calculated using the ionic strength probability value calculated by the value calculation means.

(3) As the easiness between the amino acids, further comprising a storage means storing an inter-amino acid cleavage strength probability value calculated as a probability value of the easiness between the amino acids, the evaluation value calculation means, The amino acid sequence identification device according to (1), wherein an evaluation value is calculated using an interamino acid cleavage strength probability value read from a storage means.

(4) Using the theoretical mass value of amino acids stored in the storage means, based on the mass of the peptide whose sequence is to be identified, a combination calculation that calculates a combination represented by the type and number of amino acids contained in the peptide The identification means further includes a combination of the amino acid types and the number of amino acids contained in the peptide fragment in which the candidate amino acid showing the highest evaluated evaluation value is the d + N-th amino acid, calculated by the combination calculation means. And when it is determined that there is no combination including the type and number of amino acids of the peptide fragment in these combinations, from the candidate amino acids excluding the candidate amino acids, the d + N-th amino acid in the peptide The amino acid sequence identification apparatus according to (1), wherein the identification is performed.

(5) The amino acid sequence identification device according to (1), wherein the theoretical mass value of the amino acid includes a plurality of values corresponding to peptide bond cleavage positions for one amino acid.
(6) The amino acid sequence identification device according to (1), wherein the theoretical mass value of the amino acid includes a plurality of values assuming a case where a single amino acid has chemical modification.

(7) Combination calculation for calculating a combination represented by the type and number of amino acids contained in the peptide, based on the mass of the peptide to be sequence-identified, using the theoretical mass value of the amino acid stored in the storage means. And the combination calculation means divides the mass spectrum input by the input means into a plurality of parts when the mass of the sequence identification target peptide input by the input means is greater than a specific threshold value. The amino acid sequence identification device according to (1), wherein

(8) The amino acid sequence identification device according to (4) or (7), wherein the combination calculation means performs processing using an algorithm that solves a restricted combination problem.
(9) The amino acid sequence identification device according to (1), wherein the evaluation value for the d + N-th candidate amino acid is a cumulative value of evaluation values used up to identification of peptides up to the d-th.

(10) Using the input means for inputting the mass value of the peptide obtained from the sample containing the peptide to be sequence-identified and the theoretical mass value of the amino acid read from the storage means storing the theoretical mass value of the amino acid for each amino acid. Based on the mass of the peptide whose sequence is to be identified input by the input means, the combination calculation means for calculating the combination represented by the type and number of amino acids contained in the peptide, and the combination calculation means An amino acid sequence identification device comprising an amino acid sequence identification means for identifying an amino acid sequence related to a peptide whose sequence is to be identified from among the combinations.

(11) The amino acid sequence identification device according to (10), wherein the amino acid sequence identification means identifies an amino acid sequence related to a peptide whose sequence is to be identified using a database storing amino acid sequence sequences related to known peptides.
(12) The amino acid sequence identification device according to (10), wherein the combination calculation means performs processing using an algorithm that solves a restricted combination problem.

  ADVANTAGE OF THE INVENTION According to this invention, the amino acid sequence identification apparatus, the amino acid sequence identification method, and amino acid sequence identification program which can identify an amino acid sequence at high speed and with high precision about the peptide of sequence identification object can be provided.

  Hereinafter, an amino acid sequence identification device, an amino acid sequence identification method, and an amino acid sequence identification program according to the present invention will be described in detail with reference to the drawings. The amino acid sequence identification apparatus according to the present invention is an apparatus for identifying amino acid sequences using mass spectra of identification target peptides obtained by a mass spectrometer, using amino acid sequences such as proteins, oligopeptides and polypeptides as identification targets.

<Amino acid sequence identification apparatus according to the present invention>
First, the hardware configuration of the amino acid sequence identification apparatus 1 according to the present invention will be described with reference to FIG. The amino acid sequence identification device 1 includes a CPU 101 that controls the entire system, a ROM 102 that stores a boot program, a RAM 103 that is used as a work area of the CPU 101, a hard disk (HD) 104 that stores programs, data, and the like, A hard disk drive (HDD) 105 that controls reading / writing of data with respect to the HD 104 according to the control of the CPU 101, a display 106 that displays a window related to data such as documents, images, and function information, and an external device that controls an interface between external devices and the like And a terminal 107. Each part is connected by a bus 100.

  The amino acid sequence identification apparatus 1 may be connected to a network via a communication network. In this case, the amino acid sequence identification device 1 includes an interface (I / F) that controls a network and an internal interface.

  Next, the functional configuration of the amino acid sequence identification apparatus 1 according to the present invention will be described with reference to FIG. The amino acid sequence identification device 1 includes an input processing unit 2 for inputting mass spectrum data composed of the mass-to-charge ratio and ionic strength output from the mass spectrometer 8, a candidate amino acid search processing unit 3, and an evaluation value calculation processing unit. 4 and an identification processing unit 5.

  In addition, the amino acid sequence identification device 1 has an amino acid mass table that stores the theoretical mass values of amino acids stored in the ROM 102, RAM 103, or HD 104 for each amino acid. As an example of the amino acid mass table, as shown in Table 1, a table in which 20 kinds of amino acids are associated with their masses can be exemplified.

  In Table 1, the “Average” column is the mass of each amino acid calculated from the abundance ratio of the isotopes of each amino acid, and the “Monoisotopic” column has the largest isotope abundance ratio. It is the mass of each amino acid calculated based on the mass.

  As shown in FIG. 3, it is possible to consider a plurality of positions where the peptide bond is cleaved, and it is preferable to use an amino acid mass table considering that the mass of the cleaved amino acid differs depending on the cleaved position. An example of such an amino acid mass table is shown in Table 2.

  In the amino acid mass table shown in Table 2, the column of N (a) is the mass (the mass of “a-ion”) when the peptide is cleaved at the ax line in FIG. . Similarly, the column of N (b) is the mass when the peptide is cleaved by the by line in FIG. 3 and exists on the N-terminal side (mass of “b-ion”), and the column of N (c) is FIG. The mass when the peptide is cleaved at the cz line and is present on the N-terminal side (the mass of “c-ion”). The column of C (x) is the mass when the peptide is cleaved at the ax line in FIG. 3 and exists on the C-terminal side (mass of “x-ion”), and the column of C (y) is in FIG. The mass when the peptide is cleaved at the by line and present on the C-terminal side (mass of “y-ion”), C (z) column is the C-terminal side when the peptide is cleaved at the cz line in FIG. Is the mass (the “z-ion” mass). This table shows the mass when a single amino acid appears at the end, but when a plurality of amino acids are detected, the mass of the plurality of amino acids is calculated from Table 1, and the terminal is calculated according to Table 3 below. Calculate the mass if present. The parentheses in Table 3 indicate the mass of acetylated that is likely to occur at the N-terminus.

  The amino acid mass table is not limited to those containing 20 types of amino acids as shown in Tables 1 and 2, and includes, for example, amino acid masses that take into account modifications that are considered to be susceptible to amino acids. Also good. An example of the mass of amino acids in consideration of modification is shown in Table 4, but is not limited to amino acid dehydration, deamination, and the like.

  That is, Tables 1 and 2 described above may include modified amino acids and their masses as shown in Table 4. The masses of amino acids and modified amino acids shown in Tables 1 to 4 are expressed in terms of mass to charge ratio.

  Furthermore, the amino acid sequence identification device 1 can store a value stored in the ROM 102, the RAM 103 or the HD 104, which means the ease of cutting between amino acids. As the value meaning the ease of breaking between amino acids, for example, an inter-amino acid cleavage strength probability value calculated using a statistical value of the ease of breaking between amino acids as a probability value can be used.

  The following formula is Kapp, EA, Schutz, F., Reid, GE, Eddes, JS, Moritz, RL, O'Hair, RA, Speed, TP and Simpson, RJ, Mining a Tandem Mass Spectrometry Database To Determine the Trends and Global Factors Influencing Peptide Fragmentation, Anal. Chem., 75, 6251-6264, 2003. The interamino acid cleavage strength probability value can be calculated by, for example, uniformly scaling a value calculated according to the following equation as a random variable of 0 to 1.

In the above formula, the easiness to cut in the by line shown in FIG. 3 is calculated from the number of appearances of peptide fragments consisting of b-ion or y-ion detected in the peptide, sequence, and tag. In the above formula, b _s ^{j +} means the number of occurrences of b-ion detected at j valence between amino acids (s), and y _s ^{j +} is the number of occurrences of y-ion detected at j valence between amino acids (s). Means. FIG. 4 shows an example in which the CIR _S values are summarized as statistical values of the ease of cutting for each pair of combinations selected from 20 types of amino acids.

In other words, the amino acid sequence identification device stores in the ROM 102, RAM 103 or HD 104 as a table in which CIR _s values calculated according to the above formula are associated with a pair of combinations selected from 20 types of amino acids. The formula for calculating the probability value of cleavage strength between amino acids is not limited to the above formula, and for example, the bond strength between amino acids is calculated by the ab initio molecular orbital method, and the bond strength between amino acids is calculated as a relative ratio. It can also be expressed as

  In addition, the amino acid sequence identification device 1 includes an ion intensity probability value calculation processing unit 6 that calculates the ion intensity of a mass spectrum (each peak) included in the mass spectrum data input by the input processing unit 2 as a probability value, and an identification target Combination calculation means 7 for calculating the combination represented by the type and number of amino acids contained in the peptide based on the mass of the peptide, and the mass accumulation calculation processing for calculating the accumulated mass M of the amino acid sequence identified in the identification processing unit 5 The structure provided with the part 9 may be sufficient.

  The input processing unit 2 inputs the mass spectrum data output from the mass spectrometer 8 connected via, for example, the external terminal 107. The input processing unit 2 may input mass spectrum data recorded on an information recording medium such as a CD-ROM or DVD-ROM, or the mass spectrum data stored in the server via a network. May be input.

  In the present invention, the mass spectrometer 8 for obtaining a mass spectrum to be used as an input is not particularly limited, and any mass spectrometer may be used. Here, as the mass spectrometer 8, particles such as molecules are made into gaseous ions, and these ions are moved in a vacuum electromagnetic field to separate and detect according to the mass-to-charge ratio (m / z) of the ions. It is a device to do. The mass spectrometer 8 generally includes an ion source for ionizing molecules contained in a sample, a mass analyzer for separating generated ions depending on a mass-to-charge ratio, and ions separated by the mass analyzer. An ion detection unit for detection.

  Ion sources include those that use matrix-assisted laser desorption ionization (MALDI), those that use electron spray ionization (ESI), those that use electron impact (EI), and chemical ionization (CI). The one that adopts the fast atom collision method (FAB), the one that adopts the inductively coupled plasma method (ICP), the one that adopts the thermospray, the one that adopts the atmospheric pressure chemical ionization method (APCI) can be used. . In particular, in the present invention, in order to ionize a peptide or protein, a mass spectrometer having an ion source employing MALDI or ESI which is soft ionization is preferable.

  In addition, the mass spectrometer includes a single-converging magnetic field deflection mass spectrometer, a quadrupole mass spectrometer, an ion trap mass spectrometer, a time-of-flight mass spectrometer, a Fourier transform mass spectrometer, and a double-focusing mass spectrometer. A mass spectrometer etc. can be mentioned. In addition, by connecting the mass analyzer in tandem, all ions generated in the ion source are separated by the first mass spectrometer and only the selected precursor ions are passed through to fragment, and the fragmented product ions are A tandem mass spectrometer (hereinafter referred to as MS / MS) that performs analysis in the second mass spectrometer can be given. An ion trap mass spectrometer can measure MS / MS spectra by itself. In addition, the time-of-flight mass spectrometer can measure the fragmented product ions at the same time as the precursor ions by causing post-source decomposition (PSD) in which ions exiting the ion source decompose during flight. There are also things.

  In the present invention, a mass spectrometer that can obtain a spectrum of precursor ions and fragmented product ions thereof, that is, a mass spectrometer, an ion trap mass spectrometer, and a time of flight capable of measuring a PSD spectrum. A mass spectrometer having a mass spectrometer, a tandem mass spectrometer, and the like is required.

  In addition, the candidate amino acid search processing unit 3 sets the number of amino acids (peptides) matching the set search amino acid number N to the d + 1 to d + Nth amino acids following the amino acids up to the dth in the sequence identification target peptide. Search as a candidate amino acid (candidate peptide). The candidate amino acid search processing unit 3 searches for one or more amino acids as candidate amino acids. The evaluation value calculation processing unit 4 calculates an evaluation value with respect to one or a plurality of candidate amino acids searched by the candidate amino acid search processing unit 3 using an evaluation function described later in detail. Here, the evaluation function will be described in detail later. In the mass spectrum calculated when the d + N-th amino acid is a candidate amino acid (estimated mass value) and the mass spectrum input by the input processing unit 2 The difference from the actually measured mass value (actually measured mass value) is positively evaluated when the ionic strength of the mass spectrum from which the actually measured mass value is calculated is high or when the d-th amino acid is easily cut off from the candidate amino acid. It is a function. “Positive evaluation” means that when the evaluation value calculated in accordance with the evaluation function is smaller, the difference is reflected so as to be a smaller value. The ion intensity probability value calculation processing unit 6 executes a process of scaling the mass spectrum input by the input processing unit 2 so that the ion intensity is a relative value and the relative value is uniformly distributed within the random variable. The ion intensity probability value calculated by the ion intensity probability value calculation processing unit 6 can be used for the evaluation function.

  The identification processing unit 5 identifies the d + N-th amino acid using the evaluation value calculated by the evaluation calculation processing unit 4 for the candidate amino acids searched by the candidate amino acid search processing unit 3. Although details will be described later, the identification processing unit 5 identifies the candidate amino acid having the smallest evaluation value as the d + N-th amino acid in principle, but the candidate amino acid having the smallest evaluation value is not necessarily d +. It is not necessary to identify the Nth amino acid.

  Further, the combination calculation processing unit 7 can calculate a combination represented by the type and number of amino acids contained in the peptide based on the mass of the peptide to be sequenced. The combination calculation processing unit 7 is based on, for example, the precursor ion charge mass ratio output from the MS / MS first mass spectrometer, that is, the precursor ion mass value calculated from the mass-to-charge ratio obtained from the MS spectrum. Thus, a combination represented by the type and number of amino acids contained in the precursor ion (peptide) can be calculated using the theoretical mass value of the amino acid stored in the storage means. The combination calculation processing unit 7 can calculate a combination represented by the type and number of amino acids by a process to which a solution algorithm for a limited combination problem using a so-called knapsack problem as an example is applied.

  Further, the identification processing unit 5 can identify the d + N-th amino acid by using the combination calculated by the combination calculation processing unit 7 together with the evaluation value calculated by the evaluation calculation processing unit 4.

<Amino acid sequence identification method and program according to the present invention>
Hereinafter, specific processing in the amino acid sequence identification apparatus 1 according to the present invention will be described with reference to the flowchart shown in FIG. In this example, an apparatus capable of detecting an MS / MS spectrum will be described as an example of the mass spectrometer 8. That is, a peptide fragment obtained by degrading a protein with a predetermined protein digestion enzyme is introduced into a first mass spectrometer, and the mass of the precursor ion is measured by the first mass spectrometer. Thereafter, spectral data derived from the decomposition product of a specific precursor ion is output by the second mass spectrometer.

Further, in the amino acid sequence identification device 1 described in this example, the amino acid mass table shown in Table 2 is stored in the storage means, and the inter-amino acid cleavage strength probability calculated as a statistical value of the fragility between amino acids as a probability value. A table that associates values (CIR _s values calculated according to the above formula) with combinations of a pair of amino acids is stored in the storage means. In the following example, among the peptide fragments obtained by digesting bovine serum albumin (BSA) with a known amino acid sequence with Trypsin, the amino acid of peptide “AEFVEVTK” (detection mass-to-charge ratio (divalent): 462.022) Sequence identification is illustrated below as an example. As the mass spectrometer 1, specifically, Finnigan LTQ (ThermoFinnigan) was used. In addition, Paradigm MS4 (Michrom BioResources) was used as a liquid chromatography apparatus. The obtained MS / MS spectrum is shown in FIG.

  Prior to processing in the amino acid sequence identification device 1 described in this example, the ion intensity probability value calculation processing unit 6 is expressed as the height of a peak included in the mass spectrum data input by the input processing unit 2. The ion intensity is calculated as a probability value, and an ion intensity table that associates the charge mass ratio expressed as the position of each peak with the ion intensity probability value of each peak is stored in the storage means. As an example of the ion intensity table, as shown in Table 5, the mass-to-charge ratio (“m / z” in the table) of the peak included in the mass spectrum data and the ion intensity actually measured for the peak (“ Intensity ”), relative ion intensity calculated by dividing each ion intensity by the maximum value in the ion intensity (“ Relative Intensity ”in the table), and each value of the relative ion intensity as a random variable of 0 to 1 A table in which normalized relative ionic strength (“Normalized RI” in the table) calculated by uniformly scaling can be listed.

  The scaling method can be variously changed depending on the MS / MS apparatus and measurement conditions. For scaling in this example, a scaling function represented by f (x) = 1−exp (−0.192 * x) [where x represents relative ion intensity] is used.

  The amino acid sequence identification device 1 described in this example executes the following processing after storing the ionic strength probability value table created as described above in the storage means.

  First, in step 1 (denoted as “S1” in FIG. 5 and the following steps are also the same), the mass accumulation calculation unit 9 initializes the mass accumulation value M (sets M = 0), that is, The mass accumulation calculation processing unit 9 stores M = 0 in the storage unit.

  Next, in step 2, the candidate amino acid search processing unit 3 initializes the search amino acid number N, that is, sets N = 1. That is, the candidate amino acid search processing unit 3 stores N = 1 in the storage unit.

  Next, in step 3, the mass accumulation calculation processing unit 9 adds the amino acid mass of the number of residues specified by the search amino acid number N to the mass accumulation value M, and updates the mass accumulation value M. For example, when searching for C-terminal candidate amino acids, in this step, the mass represented by the column of C (y) in the amino acid mass table shown in Table 2 (the peptide is cleaved by the by line in FIG. 3). The mass when present on the C-terminal side) is added to the cumulative mass value M. When searching for the C-terminal amino acid, the mass represented by the column of C (x) or C (z) in Table 2 may be collated with the mass spectrum data. If searching for an N-terminal amino acid, any of the masses represented by the columns N (a), N (b) and N (c) in Table 2 may be added.

  In step 3, as the mass accumulation value M, the mass accumulation value M can be updated by using all 20 types of amino acids contained in the amino acid mass table shown in Table 2, or a part of the amino acid mass table is included. It is also possible to update the mass cumulative value M by using the amino acids. In addition, when the peptide fragment to be introduced into the first mass spectrometer is obtained by a protein digestion enzyme that recognizes and digests a specific amino acid residue, amino acids that can exist at the terminal can be limited. Therefore, in this case, it is preferable to search the amino acid mass table and preferentially update the mass cumulative value M for amino acids that may exist at the terminal.

  Next, in Step 4, the candidate amino acid search processing unit 3 uses the mass spectrum data input by the input processing unit 2 or the ion intensity table constructed by the ion intensity probability value calculation processing unit 6 (for example, the table shown in Table 5). And a peak of mass spectrum data having a mass-to-charge ratio corresponding to the mass accumulation value M updated in step 3 is identified. In this step, a mass error is set in advance for the mass accumulation value M, and a peak of mass spectrum data having a mass-to-charge ratio within the mass error range can be identified. In this step, mass spectrum data may be identified for a plurality of mass accumulation values M among the mass accumulation values M updated in step 3. The amino acid indicating the mass cumulative value M for which mass spectrum data could be identified is stored in the storage means as a candidate amino acid.

  In step 4, if a mass spectrum data peak having a mass-to-charge ratio corresponding to the mass accumulation value M is identified, the process proceeds to step 5, and the case where it is not identified will be described later.

  Next, in step 5, first, the evaluation value calculation processing unit 4 calculates an evaluation value for each candidate amino acid using the evaluation function. Here, as the evaluation function, the difference between the estimated mass value and the actually measured mass value, that is, the difference between the mass accumulated value M and the mass value calculated from the mass spectrum data, the identified mass spectrum data The following formula showing a relationship that positively evaluates when the ionic strength is high and when the predetermined amino acid and the candidate amino acid are easily cut off can be used.

Score calculated by the above formula is an evaluation value. In the above formula, M _diff (i) is a value calculated as a difference between the estimated mass value and the actually measured mass value. In the above formula, I _peak (i) is a value read from the ion intensity probability value table shown in Table 5 above. In the above formula, A _CIR (i) is a value read from a table associated with CIR _s values. In the above formula, i means the search depth.

In step 5, when the candidate amino acid is located at the terminal, the ease of cutting between amino acids is evaluated equally for all candidate amino acids. That is, the evaluation value calculation processing unit 4 performs calculation with all A _CIR (i) in the evaluation function fixed to 1 when calculating the evaluation value for each candidate amino acid. The evaluation value for the candidate amino acid calculated by the evaluation value calculation processing unit 4 is stored in the storage means in association with the candidate amino acid.

  In step 5, the identification processing unit 5 then identifies the C-terminal amino acid based on the evaluation value for the candidate amino acid calculated by the evaluation value calculation processing unit 4. Specifically, the candidate amino acid showing the smallest evaluation value is identified as the C-terminal amino acid. In step 5, next, the mass accumulation calculation processing unit 9 updates the mass accumulation value M to the mass of the amino acid identified by the identification processing unit 5.

  Next, in Step 6, the identification processing unit 5 compares the mass accumulated value M updated in Step 5 with the mass value of the precursor ion. As a result of the comparison, if the identification processing unit 5 determines that the mass accumulated value M updated in step 5 and the mass value of the precursor ion match within a predetermined error range, the search is terminated, and it is determined that they do not match. If so, return to Step 3.

In step 6, the amino acid sequence of the peptide to be sequence-identified is identified by repeatedly executing steps 3 to 6 until it is determined that the mass accumulated value M and the mass value of the precursor ion match within a predetermined error range. Can do. When repeatedly executing Steps 3 to 6, in Step 3, the mass accumulation calculation processing unit 9 adds the mass represented in Table 1 to the mass accumulated value M. When Steps 3 to 6 are repeatedly executed, in Step 5, the evaluation value calculation processing unit 4 determines the ease of breaking between amino acids for each candidate amino acid when calculating the evaluation value Score using the above formula. The intensity probability value is read from a table associated with a pair of amino acids, and is calculated by substituting A _CIR (i) in the evaluation function.

  If the peak of the mass spectrum data having the mass-to-charge ratio corresponding to the mass accumulation value M updated in step 3 is not identified in step 4 that is initially performed by starting the search from the end, the process proceeds to step 7. In step 7, the number of searched amino acids N in the candidate amino acid search processing unit 3 is updated to 2. That is, even if the search is performed with the search amino acid number N set to 1, it is determined that the amino acid located at the search start terminal cannot be identified (step 4), so it is considered that the amino acid located at the search start terminal was not cleaved. Therefore, in step 7, the number of search amino acids N in the candidate amino acid search processing unit 3 is set to 2 in order to detect a dipeptide containing an amino acid located at the search start terminal in the processes after step 7.

  Similarly, when the process proceeds to step 7 when repeatedly executing steps 3 to 6, the search amino acid number N in the candidate amino acid search processing unit 3 is updated to 2. That is, it is considered that the next amino acid residue of the amino acid sequence identified by repeating Steps 3 to 6 was not cleaved alone. Therefore, in step 7, the number of search amino acids N in the candidate amino acid search processing unit 3 is set to 2 in order to detect the next search target amino acid as a dipeptide.

In step 8 after step 7, the candidate amino acid search processing unit 3, the search the number of amino acids N updated in step 7 it is determined whether a value that exceeds the threshold value N _th set in advance. That is, in step 8, the candidate amino acid search processing unit 3, the set was searched number of amino acids N against a threshold value N _th in step 7, the search the number of amino acids N and compares the threshold value N _th. If it is determined in step 8 that the search amino acid number N does not exceed the threshold value N _th , the process proceeds to step 3, and the processes in and after step 3 are similarly executed with the search amino acid number N set in step 7. On the other hand, if it is determined in step 8 that the search amino acid number N exceeds the threshold value N _th , the process proceeds to step 9. That is, when the processes of Step 7 to Step 4 (including the processes of Steps 5 and 6) are repeated, the number of searched amino acids N becomes 2 or more, and Step 7 is repeated until the number of searched amino acids N exceeds the threshold value N _th. -The process of step 4 (including the processes of steps 5 and 6) is repeated.

  When the search amino acid number N is 2 or more, in step 4, mass spectrum data (FIG. 6) in which an estimated mass value of a peptide composed of N amino acids is input by the input processing unit 2 by the candidate amino acid search processing unit 3. ) And the mass-to-charge ratio. At this time, in Step 4, among peptides consisting of N amino acids, the inter-amino acid cleavage strength probability value between amino acids calculated using the statistical values shown in FIG. It is desirable to collate with the mass-to-charge ratio of the mass spectrum data (FIG. 6) in order from the peptide with the lowest cumulative value and the smallest value at the Nth. That is, it is assumed that the amino acids up to the (N-1) th are not detected, that the amino acids up to the (N-1) th are difficult to cleave (hard to cut), and easy to cleave (easy to cut) at the Nth. I mean.

  On the other hand, in step 9, the mass accumulation calculation processing unit 9 returns the mass accumulation value M to the mass value of the amino acid sequence up to the position where the search should be performed again. That is, the mass accumulation calculation processing unit 9 updates the mass accumulation value M to a value obtained by subtracting the mass of the amino acid on the terminal side opposite to the search start terminal. After step 9, in step 2, the search amino acid number N is initialized to 1. Thereafter, in step 3, the mass cumulative calculation processing unit 9 updates the mass cumulative value M as described above, but amino acids other than the amino acid subtracted in step 9 are set as candidate amino acids, and the mass cumulative value M is set for these candidate amino acids. Update. In step 3 executed after step 9, candidate amino acids may be selected by searching the evaluation values calculated so far stored in the storage means.

  As described above, by executing the processing according to the flowchart shown in FIG. 5, the amino acid sequence identification device 1 identifies the amino acid sequence of a specific precursor ion using the spectrum data output from the mass spectrometer 8. Can do. In step 5, the amino acid sequence identification apparatus 1 calculates the evaluation value Score by reflecting the ease of cutting between adjacent amino acids for each candidate amino acid and the ion intensity in the mass spectrum data for each candidate amino acid. Further, in step 5, the amino acid to be identified is identified based on the evaluation value Score for each candidate amino acid. In general, for example, isoleucine and leucine have the same mass, and therefore cannot be determined only from the mass-to-charge ratio of the mass spectrum data. However, according to Step 5 described above, by comparing the evaluation value when the candidate amino acid is isoleucine and the evaluation value when the candidate amino acid is leucine, identifying a more likely amino acid sequence Can do.

  As described above, by executing the processing according to the flowchart shown in FIG. 5, the amino acid sequence identification device 1 allows the amino acid sequence of the peptide to be sequence-identified without using an existing database storing the amino acid sequence and the like. Can be determined.

  In particular, the process according to the flowchart shown in FIG. 5 calculates the mass of the precursor ion in advance from the charge mass ratio of the precursor ion, and from the mass, the type of amino acid contained in the precursor ion (sequence identification target peptide) and The process may include a step of calculating a combination represented by the number. This processing is performed by the combination calculation processing unit 7. The combination calculation processing unit 7 uses the amino acid mass tables shown in Tables 1 to 4 and combinations represented by the type and number of amino acids by applying an algorithm that solves the knapsack problem, which is one of the solutions for the limited combination problem. Can be calculated. An algorithm for solving the knapsack problem is known as the problem of selecting a combination of items that maximizes the total value when a knapsack of different sizes and values is packed into a knapsack of a specific capacity. It has been. The combination calculation processing unit 7 considers the capacity of the knapsack as the mass of the precursor ion, and regards the packed bag as the mass and the number of each amino acid, and solves this problem by the solution of the “knapsack problem”, that is, the type of amino acid. And a combination of the numbers is calculated. The combination calculation processing unit 7 sets an error with respect to the mass of the precursor ion corresponding to the capacity of the knapsack, and lists all combinations of amino acid types and numbers that fall within the mass obtained by adding the error.

  Although the explanation has been given by taking the knapsack problem as an example of the computer algorithm, the combination calculation processing unit 7 applies all the combinations of amino acids by applying a computer algorithm that solves the entire solution search problem that satisfies a given constraint condition. Can also be listed. The combination of amino acid type and number calculated by the combination processing unit 7 is stored in a storage unit such as the RAM 103, for example.

  When the combination calculation processing unit 7 calculates the combination of the type and number of amino acids, in step 4, the candidate amino acid search processing unit 3 identifies a peak of mass spectrum data having a mass-to-charge ratio corresponding to the mass accumulation value M. Then, when searching for candidate amino acids, the data of the above combination stored in the storage means is collated to identify candidate amino acids that can satisfy the combination. Thereby, candidate amino acids to be searched in Step 4 can be narrowed down, so that the processing speed can be improved and the accuracy of the amino acid sequence identification result can be improved.

  In addition, the step in which the combination calculation processing unit 7 calculates the combination of the type and number of amino acids may be executed prior to Step 1 or may be executed in parallel with Steps 1 to 3. However, this step is executed before the candidate amino acid search processing unit 3 identifies a specific amino acid from a plurality of candidate amino acids in step 4.

  By the way, as shown in FIG. 7, the amino acid sequence identification device 10 according to the present invention identifies an amino acid sequence related to a sequence identification target peptide from the combinations calculated by the combination calculation processing unit 7 and the combination calculation means 7. And an amino acid sequence identification unit 11 that performs the above-described process. The combination calculation processing unit 7 is the same as the combination calculation processing unit 7 shown in FIG. 2 described above, and detailed description thereof is omitted. The amino acid sequence identification processing unit 10 may be composed of the candidate amino acid search processing unit 3, the evaluation value calculation processing unit 4, the identification processing unit 5 and the mass accumulation calculation processing unit 9 shown in FIG. Alternatively, the peptide sequence tag method may be executed.

  The amino acid sequence identification device 10 configured as described above can identify the amino acid sequence of the peptide (precursor ion) to be sequenced according to the flowchart shown in FIG. Also in this example, an apparatus capable of detecting an MS / MS spectrum will be described as an example of the mass spectrometer 8.

  First, in step 11 (denoted as “S11” in FIG. 8. The following steps are also the same), the mass-to-charge ratio of a specific precursor ion output from the first mass spectrometer of the mass analyzer 8 is input processing unit. 2. The mass of the precursor ions is calculated from the mass-to-charge ratio of the entered precursor ions.

  The mass of the precursor ion measured by the first mass spectrometer is the peptide that is not charged in the input processing unit 2 using the mass-to-charge ratio of the precursor ion measured by the first mass spectrometer of the mass spectrometer 8. The mass of is calculated. Specifically, the mass of the peptide can be obtained by subtracting the mass of the proton from the mass-to-charge ratio of the precursor ion and calculating the product with the charge.

Next, in step 12, the calculated precursor ion mass, against the threshold value M _th stored in the storage means of the ROM102 etc. and preset, the precursor ion mass is determined greater than M _th. If it is determined that the precursor ion mass is greater than M _th , the process proceeds to step 13, and if it is determined that the precursor ion mass is less than M _th , the process proceeds to step 14.

Here, the threshold value M _th can be determined as appropriate in consideration of a large burden on computer resources. That is, if the computer resource increases, a larger value can be set as the threshold value M _th . For example, 492 types if you set the M _th divalent mass to charge ratio 462.022 of a precursor, i.e. mass 922.028 used in Figure 9-1 to 9-5 described below, as shown in at least FIG. 9-1 to 9-5 the combinations of amino acids needs to be stored in the RAM 103, as long obtained sufficiently processable computer resources the combination of amino acids to be stored therein and sufficient capacity RAM 103, to be capable of setting the M _th more large Become.

Next, in step 13, the spectrum data derived from the precursor ion decomposition product input in step 11 is divided into a plurality (for example, divided into two) at a predetermined position. Here, the position to be divided is optimally selected from the peak position indicating the mass-to-charge ratio in the input spectrum data and the peak height indicating the ion intensity. For example, the mass of the sequence-identified peptide after the division is divided by the peak position does not exceed the threshold value M _th.

  Next, in step 14, the combination calculation processing unit 7 calculates the precursor ion mass measured by the first mass spectrometer, or the amino acid contained in the sequence identification target peptide from the divided mass divided in step 13. A combination represented by type and number (hereinafter referred to as a candidate amino acid set) is calculated.

  In the calculation of the candidate amino acid set in the combination calculation processing unit 7, the candidate amino acid set is calculated using a computer algorithm assuming that the mass has given an arbitrary error to the mass of the precursor. As the computer algorithm, for example, an algorithm used when solving a so-called knapsack problem can be applied. As an arbitrary error, ± 0.005 can be set with respect to the mass-to-charge ratio of the precursor ion. More specifically, when the mass-to-charge ratio of the precursor ion is 462.022 (charge: divalent), the combination of the types and number of amino acids constituting the peptide corresponding to the mass from mass-to-charge ratio 462.017 to 462.027 (492 ways) ) Are shown in FIGS. 9-1 to 9-5.

  In addition, in the case where the peptide for sequence identification is obtained by processing of a protein digestive enzyme or the like, it is possible to narrow down combinations of calculation results by considering the recognition sequence of the digestive enzyme. 1 to 9-5 are the results in consideration of this.

  The knapsack problem is known as a problem of selecting a combination of items that maximizes the total value when several types of packages having different sizes and values are packed into a knapsack of a specific capacity. The combination calculation processing unit 7 calculates the type and number of amino acids constituting the precursor ion by regarding the knapsack capacity as the mass of the precursor and associating the packed goods with the mass and number of each amino acid. When using the knapsack problem solution in the processing of the combination processing unit 7, unlike the actual knapsack problem solution, an error is set with respect to the mass of the precursor ion corresponding to the knapsack capacity, and within the capacity error (mass All combinations of packages (amino acids) that fall within (within error) are listed.

  In addition, the combination calculation processing unit 7 is not limited to the process of applying the knapsack problem solving algorithm, but performs the process of applying a computer algorithm that solves the entire solution search problem that satisfies the given constraint condition, thereby obtaining the precursor ion. It is also possible to list all combinations of types and numbers of amino acids constituting the amino acid. The combination of amino acid type and number calculated by the combination processing unit 7 is stored in a storage unit such as the RAM 103, for example.

  Next, in step 15, a sequence identification method to be applied is selected. When the combination of the type and number of amino acids calculated in step 14 and stored in the storage means is used (method 1), the process proceeds to step 16, and a method for identifying the amino acid sequence of the sequence identification target peptide is as described above. Processing according to the flowchart shown in FIG. Moreover, when using the combination of the kind and number of amino acids stored in the storage means, the so-called peptide sequence tag method can be applied to identify the amino acid sequence of the sequence identification target peptide (Method 2).

  In the peptide sequence tag method, first, in step 17, since a storage means storing a combination of amino acid types and numbers is used as a database, a permutation of all amino acids assumed from the combination is calculated. At the time of permutation calculation, amino acid sequences are listed in consideration of having a predetermined amino acid as a terminal in consideration of digestive enzymes. For example, taking the combination “AE (2) FKTV (2)” as an example among the combinations (492) shown in FIGS. 9-1 to 9-5, 1,260 kinds of amino acid sequences including the following are obtained. . In the peptide sequence tag method, these sequences are regarded as an amino acid sequence database and used for amino acid sequence identification.

  Next, in step 16, the amino acid sequence can be identified by the peptide sequence tag method by using the permutation of the amino acid sequence thus constructed as a database. Here, the peptide sequence tag method means that after a specific precursor ion is selected in a mass spectrometer, a product ion spectrum (MS / MS spectrum) obtained from the selected precursor ion is used. This means a method of performing a database search using the calculated amino acid sequence as a database. Since peptide ion cleavage by collision occurs at almost a fixed position (mainly the main chain: see FIG. 2), the product ion mass of the amino acid sequence in step 17 can be predicted. Therefore, in the peptide sequence tag method in this example, the mass of the product ion assumed from the amino acid sequence calculated in step 17 is compared with the mass value indicated by the MS / MS spectrum.

Specifically, based on the results of mass spectrometry of peptide fragment “AEFVEVTK” (detected mass-to-charge ratio (divalent): 462.022) obtained by digesting bovine serum albumin (BSA) with known amino acid sequence with Trypsin Amino acid sequence identification was performed. When the amino acid sequence obtained in step 17 is obtained as a mass analysis result, the mass-to-charge ratio (Average) of each ion, that is, the mass-to-charge ratio of the product ions and the ms / ms peak list shown in Table 5 are ± 0.5 The following amino acid sequences such as 'AEFVEVTK' were identified by comparing the mass-to-charge ratios that gave the above error. As an example, “AVFVEVTK” in the peptide sequence calculated in Step 17 is b-ion, y-ion, and the mass when they are detected as multivalent ions, dehydration, and deamination. Table 7 shows the mass values that matched with an error of ± 0.5.

  Of the amino acid sequences calculated in step 17, “AEFVEVTK” is recognized as a sequence identification result by being consistent with the MS / MSS peak list in Table 5 over the entire sequence. In Table 7, y and b represent each cleavage ion series, ++ represents a divalent ion, * represents deamination, and O represents a mass during dehydration.

It is a schematic block diagram for demonstrating the hardware constitutions of the amino acid sequence identification device which concerns on this invention. It is a schematic block diagram for demonstrating the functional structure of the amino acid sequence identification device 1 which concerns on this invention. It is a figure for demonstrating the molecular structure of the cleavage position in a peptide bond, and the cleavage amino acid produced | generated after cleavage. It is a characteristic view which shows the easiness of cutting between a pair of amino acids. It is a flowchart which shows an example of the specific process to which the amino acid sequence identification method and program which concern on this invention are applied. Among the peptide fragments obtained by digesting bovine serum albumin (BSA) with Trypsin, a characteristic showing an MS / MS spectrum obtained by using the peptide “AEFVEVTK” (detected mass-to-charge ratio (divalent): 462.022) as a precursor ion FIG. It is a schematic block diagram for demonstrating the functional structure of the amino acid sequence identification apparatus 10 which concerns on this invention. It is a flowchart which shows the other example of the specific process to which the amino acid sequence identification method and program which concern on this invention are applied. It is a figure which shows the result of having calculated the combination (492 way) of the kind and number of amino acids which comprise the peptide corresponding to the mass from mass to charge ratio 462.017 to 462.027. It is a figure which shows the result of having calculated the combination (492 way) of the kind and number of amino acids which comprise the peptide corresponding to the mass from mass to charge ratio 462.017 to 462.027. It is a figure which shows the result of having calculated the combination (492 way) of the kind and number of amino acids which comprise the peptide corresponding to the mass from mass to charge ratio 462.017 to 462.027. It is a figure which shows the result of having calculated the combination (492 way) of the kind and number of amino acids which comprise the peptide corresponding to the mass from mass to charge ratio 462.017 to 462.027. It is a figure which shows the result of having calculated the combination (492 way) of the kind and number of amino acids which comprise the peptide corresponding to the mass from mass to charge ratio 462.017 to 462.027.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Amino acid sequence identification device, 2 ... Input processing part, 3 ... Candidate amino acid search processing part, 4 ... Evaluation value calculation processing part, 5 ... Identification processing part

Claims (9)

An input means for inputting a mass spectrum of a product ion consisting of a mass-to-charge ratio and an ionic strength and a mass of a precursor ion obtained from a sample containing a peptide to be sequenced;
From the storage means that stores the theoretical mass value of amino acids for each amino acid, the estimated mass value is calculated based on the amino acid sequence of peptide fragments up to the d-th (d is an integer of 0 or more), and the calculated mass value of the amino acid is Calculate the estimated mass values of the peptide fragments up to d + Nth (N is the number of searched amino acids, an integer of 1 or more) obtained by adding the theoretical mass values, respectively. Calculate the difference between the estimated mass value and the measured mass value calculated from the mass-to-charge ratio of the specific mass spectrum among the mass spectra input from the input means, and the calculated difference is within a predetermined range. A candidate amino acid search means for identifying and using the N amino acids used to calculate the identified difference as a candidate amino acid or a candidate amino acid sequence of the d + 1 to d + N-th amino acids;
For each candidate amino acid or each candidate amino acid sequence searched by the candidate amino acid search means, for the difference identified above , a value indicating the height of ionic strength of the specific mass spectrum and the d-th amino acid and the candidate amino acid or candidate An evaluation value calculating means for calculating an evaluation value using an evaluation function including a value indicating the ease of cutting between amino acid sequences as a variable ;
By identifying one candidate amino acid from the candidate amino acids using the evaluation value obtained or identifying one candidate amino acid sequence from the candidate amino acid sequence, the amino acid sequence in the peptide fragment up to d + Nth in the peptide a identification means for identifying,
The candidate amino acid search means, the evaluation value calculation means, and the identification means identify the amino acid sequence from the terminal to the N-th in the sequence identification target peptide as d = 0, and then calculate based on the identified amino acid sequence An amino acid sequence identification apparatus for identifying an amino acid sequence in peptide fragments up to d + Nth by using an estimated mass value as a dth peptide fragment . Ion intensity probability value calculating means for scaling the ion intensity included in the mass spectrum input by the input means so as to be uniformly distributed within the random variable,
2. The amino acid sequence identification apparatus according to claim 1, wherein the evaluation value calculating means calculates an evaluation value using the ion intensity probability value calculated by the ion intensity probability value calculating means. As a value indicating the ease of cutting, further comprising a storage means for storing an inter-amino acid cleavage strength probability value calculated as a probability value of a statistical value of the ease of cutting between amino acids,
2. The amino acid sequence identification apparatus according to claim 1, wherein the evaluation value calculation means calculates an evaluation value using an interamino acid cleavage strength probability value read from the storage means. Combination calculation means for calculating a combination represented by the type and number of amino acids contained in the peptide based on the mass of the peptide to be sequence-identified using the theoretical mass value of the amino acid stored in the storage means. Prepared,
The identification means collates the type and number of amino acids contained in the peptide fragment with the candidate amino acid showing the highest evaluation value as the d + N-th amino acid with the combination calculated by the combination calculation means, and When it is determined that there is no combination including the type and number of amino acids of the peptide fragment in the combination, the d + N-th amino acid in the peptide is identified from the candidate amino acids excluding the candidate amino acid. The amino acid sequence identification device according to claim 1.   2. The amino acid sequence identification apparatus according to claim 1, wherein the theoretical mass value of the amino acid includes a plurality of values corresponding to peptide bond cleavage positions with respect to one amino acid.   2. The amino acid sequence identification apparatus according to claim 1, wherein the theoretical mass value of the amino acid includes a plurality of values assuming a case where chemical modification is made with respect to one amino acid. Combination calculation means for calculating a combination represented by the type and number of amino acids contained in the peptide based on the mass of the peptide to be sequence-identified using the theoretical mass value of the amino acid stored in the storage means. Prepared,
When the mass of the peptide for sequence identification input by the input unit is larger than a specific threshold, the combination calculation unit divides the mass spectrum input by the input unit into a plurality of processes, The amino acid sequence identification device according to claim 1.   The amino acid sequence identification apparatus according to claim 4 or 7, wherein the combination calculation means performs processing using an algorithm for solving a restricted combination problem. 2. The amino acid sequence identification device according to claim 1, wherein the evaluation value for the d + Nth candidate amino acid or the candidate amino acid sequence is a cumulative value of evaluation values used up to identification of peptides up to the dth. .

Images