Method, system and program for analyzing mass spectrometoric data

Abstract

Provided is a mass spectrometric data analyzing method for deducing the structure of an unknown substance from data obtained by an MS.sup.n analysis, in which a structural candidate having a high degree of freedom for covering a structural change of the known substance can be created. In the mass spectrometric data analyzing method according to the present invention, a candidate of the partial structure of a known substance which is structurally similar to an unknown substance as the target of deduction is created by eliminating a part of the structure of the known substance (Step S1). Previously given candidates of known additional structural parts are individually added to each candidate of the partial structure of the known substance, thus forming various combinations (Step S5). All the structural formulae that can be derived from each combination are created as the structural candidates of the unknown substance (Step S6).

Claims

1. A mass spectrometric data analyzing method for deducing a structure of an unknown substance from a result of an MS.sup.n analysis of the unknown substance (n≥2) and structural information of a known substance similar to the unknown substance, comprising steps of: running an MS.sup.n analysis of the unknown substance with a mass spectrometer to generate an MS.sup.n spectrum of the unknown substance; detecting a fragment ion peak and/or a neutral loss from the MS.sup.n spectrum of the unknown substance, generated by the mass spectrometer; generating candidates of a partial structure of the known substance by eliminating a part of a structure of the known substance from a structural formula of the known substance, without relying on a predetermined chemical or metabolic reaction of the known substance; storing known structural parts as candidates of an additional structural part; creating combinations of the candidates of the partial structure of the known substance and the known candidates of the additional structural part without relying on a predetermined chemical or metabolic reaction of the known substance; selecting, as a candidate of combination, a combination whose total mass matches with a mass of the unknown substance obtained from the MS.sup.n analysis by running the mass spectrometer; and creating, as a structural candidate of the unknown substance, a structure that can be derived from each of the candidates of combination.

2. The mass spectrometric data analyzing method according to claim 1, wherein said generating candidates further comprises steps of: detecting a fragment ion peak and/or a neutral loss common to both an MS.sup.n spectrum of the known substance and an MS.sup.n spectrum of the unknown substance; and narrowing the candidates of the partial structure of the known substance down to a candidate or candidates which include the partial structure of the known substance corresponding to the common fragment ion peak and/or the neutral loss.

3. The mass spectrometric data analyzing method according to claim 1, wherein said creating combinations comprises deriving a composition formula from the MS.sup.n spectrum of the unknown spectrum and selecting, from the combinations of the candidates of the partial structure of the known substance and the candidates of the additional structural parts, a combination, as a candidate of combination, whose composition formula matches with the derived composition formula.

4. The mass spectrometric data analyzing method according to claim 1, further comprising a step of selecting a candidate of the partial structure of the known substance and/or a candidate of the additional structural part to be used in said creating combinations.

5. The mass spectrometric data analyzing method according to claim 1, further comprising steps of: deducing, for the structural candidates created in said creating a structure, a fragment ion peak originating from each of the structural candidates; and comparing the deduced fragment ion peak to a fragment ion peak of the unknown substance, so as to rank the structural candidates in descending order of their probability of being the structure of the unknown substance.

6. The mass spectrometric data analyzing method, according to claim 1, wherein the step of generating candidates of a partial structure of the known substance by eliminating a part of a structure of the known substance from a structural formula of the known substance, does not rely on a predetermined chemical or metabolic reaction of the known substance so as to deal with a reaction that causes an unknown structural change of the unknown substance.

7. A mass spectrometric data analyzing system for deducing a structure of an unknown substance from a result of an MS.sup.n analysis of the unknown substance (n≥2) and structural information of a known substance similar to the unknown substance, comprising: a known substance memory in which a data of a structural formula of the known substance is stored; an additional structural part memory in which a data of a structural formula of a candidate of an additional structural part is stored; a partial structure candidate generator for generating a candidate of a partial structure of the known substance, using the structural formula of the known substance retrieved from the known substance memory, without relying on a predetermined chemical or metabolic reaction of the known substance; a combination candidate creator for creating combinations of the candidates of the partial structure of the known substance and the candidates of an additional structural part, without relying on a predetermined chemical or metabolic reaction of the known substance, and for selecting, as a candidate of combination, a combination whose total mass matches with a mass of the unknown substance; and a structural candidate creator for creating, as a structural candidate of the unknown substance, a structure that can be derived from each of the candidates of combination.

8. The mass spectrometric data analyzing system according to claim 7, wherein: the known substance memory further holds a data of an MS.sup.n spectrum of the known substance and a data of the partial structure of the known substance corresponding to a fragment ion peak and/or a neutral loss on the MS.sup.n spectrum; and the partial structure candidate creator further comprises: a detector for detecting a fragment ion peak and/or a neutral loss common to both an MS.sup.n spectrum of the known substance and an MS.sup.n spectrum of the unknown substance; and a structural candidate narrowing section for narrowing the candidates of the partial structure of the known substance down to a candidate or candidates which include the partial structure of the known substance corresponding to the common fragment ion peak and/or the neutral loss.

9. The mass spectrometric data analyzing system according to claim 7, wherein the combination candidate creator derives a composition formula from an MS.sup.n spectrum of the unknown spectrum, and selects, from the combinations of the candidates of the partial structure of the known substance and the candidates of the additional structural parts, a combination, as a candidate of combination, whose composition formula matches with the derived composition formula.

10. The mass spectrometric data analyzing system according to claim 7, wherein the combination candidate creator further comprises a candidate selector for allowing a user to select a candidate of the partial structure of the known substance and/or a candidate of the additional structural part to be used in the combination candidate creator.

11. The mass spectrometric data analyzing system according to claim 7, further comprising a rank determiner for deducing, for a plurality of structural candidates created by the structural candidate creator, a fragment ion peak originating from each of the structural candidates, and for comparing the deduced fragment ion peak to a fragment ion peak of the unknown substance, so as to rank the structural candidates in descending order of their probability.

12. A mass spectrometric data analyzing program for making a computer function as the mass spectrometric data analyzing system according to claim 7.

13. The mass spectrometric data analyzing system, according to claim 7, wherein the partial structure candidate generator for generating a candidate of a partial structure of the known substance, using the structural formula of the known substance retrieved from the known substance memory, does not rely on a predetermined chemical or metabolic reaction of the known substance so as to deal with a reaction that causes an unknown structural change of the unknown substance.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is schematic configuration diagram showing one example of the system for carrying out the mass spectrometric data analyzing method according to the present invention.

(2) FIG. 2 is a flowchart showing a procedure of the structure analysis of an unknown substance by one embodiment of the mass spectrometric data analyzing method according to the present invention.

(3) FIGS. 3A-3C are diagrams showing an example of the combination of a candidate of the partial structure of a known substance and a candidate of the additional structural part to be added to it.

(4) FIG. 4 is a structural formula of nefazodone as the known substance (metabolic precursor).

(5) FIG. 5A is an MS.sup.2 spectrum of nefazodone as the known substance (metabolic precursor) and FIG. 5B is an MS.sup.2 spectrum of an oxygen metabolite of nefazodone as the unknown substance (a product of metabolism).

(6) FIGS. 6A-6D are structural formulae of the last candidates of the partial structure of nefazodone.

(7) FIG. 7 is a set of structural formulae of the structural candidates of the unknown substance created by the mass spectrometric data analyzing method of the present embodiment.

BEST MODE FOR CARRYING OUT THE INVENTION

(8) One embodiment of a mass spectrometric data analyzing system for carrying out the mass spectrometric data analyzing method according to the present invention (which is hereinafter simply called the “data analyzing system”) is hereinafter described with reference to the drawings. The following descriptions deal with the case of using a result of an analysis by a liquid chromatograph mass spectrometer. The same discussions also holds true in the case of using a gas chromatograph mass spectrometer or other types of chromatograph mass spectrometers, as well as in the case of using a mass spectrometer into which a sample is directly introduced.

(9) FIG. 1 shows the configuration of the data analyzing system 10 of the present embodiment. The data analyzing system 10, which is used for processing detection data sent from a liquid chromatograph mass spectrometer (LC/MS) 20, includes a central controller 11, a spectrum data creator 12, an analytical processor 13, a measurement data memory 14, a reference data memory 15 and an additional structural part memory 16. The measurement data memory 14, the reference data memory 15 and the additional structural part memory 16 are connected to the central controller 11. An input unit 30 and a display unit 40 are connected to the central controller 11. The central controller 11 also has the function of controlling the operations of each section of the LC/MS 20.

(10) The central controller 11, the spectrum data creator 12 and the analytical processor 13 can be created by a central processing unit (CPU) which operates according to a computer program. For the measurement data memory 14, a readable and writable storage device can be used, such as a hard disk (HD) or magneto-optical disc (MO). The hard disk or similar type of device can also be used for the reference data memory 15 and the additional structural part memory 16, although a read-only device, such as a CD-ROM, may be used as them in a system which requires no recording of data in the reference data memory 15 or the additional structural part memory 16. The reference data memory 15 holds data relating to various kinds of compounds (e.g. the structural formula, composition formula, MS.sup.n spectrum data, and information relating to the partial structure corresponding to each fragment ion peak in the MS.sup.n spectrum data). The additional structural part memory 16 holds information relating to the candidates of the additional structural part to be added to the partial structure of the known substance (e.g. the structural formula, composition formula, mass, and number of charges).

(11) Although the measurement data memory 14, the reference data memory 15 and the additional structural part memory 16 in FIG. 1 are independently provided, these memory units may be created by logically partitioning a single storage device (such as a hard disk). For the input unit 3, a keyboard, mouse and/or other input devices can be used. For the display unit 40, a cathode ray tube (CRT), printer and/or other output devices can be used.

(12) The data analyzing system 10 can be embodied as a controlling and analyzing unit dedicated to an LC/MS system. Alternatively, it may be embodied by installing, in a personal computer or similar multi-purpose computer, a program for performing an analytical method according to the present invention.

(13) The LS/MS 20 is composed of a liquid chromatographic apparatus having a column for temporally separating a mixed sample into components and a mass spectrometer capable of an MS.sup.n analysis (at least an MS.sup.2 analysis). Examples of such a mass spectrometer include a triple quadrupole mass spectrometer and an ion trap mass spectrometer. The sample components eluted from the column of the chromatographic apparatus are sequentially ionized and subjected to an MS.sup.n analysis by the mass spectrometer. In this analysis, an ion having an appropriate mass is automatically selected as a precursor ion from the ions originating from the sample components, and the precursor ion is dissociated into fragment ions, which are then made to undergo the mass separation and detection. It is also possible to repeat the selection, dissociation and detection of the ions a plurality of times as needed.

(14) During the measurement of the sample, detection data (digitized detection signals) based on an electric current corresponding to the number of ions detected for each mass are sent from the LC/MS 20 to the data analyzing system 10. The spectrum data creator 12 processes these detection data according to a predetermined algorithm, to compute, at each stage of the MS.sup.n analysis, an MS.sup.n spectrum data consisting of a mass paired with a peak intensity (relative signal intensity) observed at the mass. The obtained data are stored in the measurement data memory 14.

(15) The foregoing descriptions related to the operations of the LC/MS 20 and the data analyzing system 10 during the measurement. The following descriptions, with reference to the flowchart of FIG. 2, explain a procedure of the structural analysis according to the present embodiment using MS.sup.n spectrum data obtained for an unknown substance having a predetermined mass and MS.sup.n spectrum data of a known substance which is structurally similar to the unknown substance, the latter data being stored in the reference data memory 15.

(16) In the present embodiment, the structural analysis is performed on the assumption that a known substance which is structurally similar to the unknown substance is previously known. For example, if the unknown substance to be analyzed is a product of metabolism or chemical change, such as a drug metabolite, and if the precursor used for that reaction is previously known, this precursor can be used as the known substance in the present embodiment. It is also possible to perform a multivariate analysis or similar computation using MS.sup.n spectrum data of the unknown substance and those of the various compounds stored in the reference data memory 15, so as to find a compound considerably similar to the unknown substance in terms of MS.sup.n spectrum data and select that compound as the known substance for the present embodiment.

(17) For the known substance which is structurally similar to the unknown substance, the analytical processor 13 performs a data processing as follows: Initially, the structural formula of the known substance is retrieved from the reference data memory 15, and candidates of the partial structure are exhaustively created by eliminating each possible part of the structural formula (Step S1). Subsequently, with reference to the MS.sup.n spectrum data of the unknown substance stored in the measurement data memory 14 and those of the known substance stored in the reference data memory 15, a fragment ion peak and/or neutral loss common to the two sets of data is detected (Step S2). Then, the candidates of the partial structure of the known substance created in Step S1 are narrowed down to those which include a partial structure of the known substance corresponding to the fragment ion peak and/or the neutral loss detected in Step S2 (Step S3).

(18) Subsequently, the composition formula of the unknown substance is deduced from its MS.sup.n spectrum data (Step S4). Meanwhile, the candidates of the additional structural part stored in the additional structural part memory 16 are individually added to each candidate of the partial structure of the known substance selected in Step S3, thus forming various combinations, and a combination which matches with the composition formula deduced in Step S4 is created as a candidate of combination (Step S5). Then, all the structural formulae that can be derived from each candidate of combination are created as the structural candidates of the unknown substance (Step S6). FIGS. 3A-3C show one example, in which a hydroxyl group (FIG. 3B) as a candidate of the additional structural part is added to a candidate of the partial structure of a known substance with a single-bond chain at both ends (FIG. 3A). In this example, the bonding of the hydroxyl group can occur at either of the two single-bond chains of the candidate of the partial structure of the known substance (FIG. 3A). Accordingly, two structural formulae can be derived from this candidate of combination. In this manner, there may be a plurality of structural formulae derived from one candidate of combination in Step S6. Conversely, it is possible that no structural formula can be found for some reasons, such as a discrepancy in the number of charges.

(19) Subsequently, for each of the structural candidates of the unknown substance created in Step S6, a fragment ion peak which is expected to be detected in an MS.sup.n analysis is deduced, and the degree of matching of this peak with a fragment ion peak existing in the MS.sup.n spectrum of the unknown substance is numerically expressed as the level of probability. Then, the structural candidates is ranked in descending order of probability (Step S7), and the result is shown on the display unit 40 (Step S8).

(20) Thus far, the mass spectrometric data analyzing method according to the present invention has been described by means of the embodiment. It should be noted that the present invention is not limited to the previous embodiment but can be changed in various ways within the spirit of the present invention.

(21) For example, the mass spectrometric data analyzing method according to the present invention can be performed without the processes of Steps S2, S3 and S7. Furthermore, the process of narrowing down the candidates by the composition formula in Steps S4 and S5 may be achieved by selecting each and every combination whose total mass matches with the mass of the unknown substance.

(22) In the flowchart of FIG. 2, it is assumed that the data analyzing system 10 automatically performs all the steps. However, for example, it is also possible to allow skilled persons to select candidates to be used for the structural deduction from the created candidates of the partial structure of the known substance and the previously given candidates of the additional structural part, based on their own experience. By this method, the processing time for the structural deduction can be reduced if it is to some extent known beforehand which partial structure will remain and which structural part will be added.

(23) FIGS. 4-7 show the result of an experiment of the structural analysis of an unknown substance. In the experiment, an oxygen metabolite of nefazodone was taken as the unknown substance (although its structure was actually known) and nefazodone (the metabolic precursor) as the known substance. The aforementioned selection of candidates by users was performed before the structural analysis was executed.

(24) FIG. 4 is the structural formula of nefazodone as the known substance (metabolic precursor). FIG. 5A is an MS.sup.2 spectrum data retrieved from the reference data memory 15. FIG. 5B is an MS.sup.2 spectrum data obtained by an MS.sup.2 analysis of the oxygen metabolite of nefazodone as the unknown substance (a product of metabolism). FIG. 6A-6D are the structural formulae of the last candidates of the partial structure of nefazodone selected by a user after the candidates of the partial structure created from the structural formula of nefazodone shown in FIG. 4 had been narrowed down by the common neutral losses detected in FIGS. 5A and 5B. FIG. 7 is a set of structural candidates of the unknown substance created by the mass spectrometric data analyzing method of the present embodiment, with the candidates being ranked in order of evaluation value. The evaluation value was obtained by dividing the sum of the intensities of the peaks to which the partial structure was assigned by the sum of the intensities of all the peaks.

(25) As shown in FIG. 7, ten structural candidates of the unknown substance were created in the experiment, and it was confirmed that the structural formula of the oxygen metabolite of nefazodone as the target of deduction was included in them (the second structural formula from the left in the upper row in FIG. 7). The time required for the entire process was 10.7 seconds. Thus, by the mass spectrometric data analyzing method of the present embodiment, the correct structural candidate can be obtained in a practically acceptable processing time even in the structural analysis of a substance having a wide variety of structural change patterns, as in the case of drug metabolites.

EXPLANATION OF NUMERALS

(26) 10 . . . Data Analyzing System 11 . . . Central Controller 12 . . . Spectrum Data creator 13 . . . Analytical processor 14 . . . Measurement Data Memory 15 . . . Reference Data Memory 16 . . . Additional Structural Part Memory 20 . . . LC/MS 30 . . . Input Unit 40 . . . Display Unit