Whole Protein Dissociation in a Quadrupole Ion Trap: Identification of

Ravi Amunugama, Jason M. Hogan, Kelly A. Newton, and Scott A. McLuckey*. Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, ...
0 downloads 0 Views 121KB Size
Anal. Chem. 2004, 76, 720-727

Whole Protein Dissociation in a Quadrupole Ion Trap: Identification of an a Priori Unknown Modified Protein Ravi Amunugama, Jason M. Hogan, Kelly A. Newton, and Scott A. McLuckey*

Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana 47907-2084

A protein mixture derived from a whole cell lysate fraction of Saccharomyces cerevisiae, which contains roughly 19 proteins, has been analyzed to identify an a priori unknown modified protein using a quadrupole ion trap tandem mass spectrometer. Collection of the experimental data was facilitated by collision-induced dissociation and ion/ion proton-transfer reactions in multistage mass spectrometry procedures. Ion/ion reactions were used to manipulate charge states of both parent ions and product ions for the purpose of concentrating charge into the parent ion of interest and to reduce the product ion charge states for determination of product ion mass and abundance. The identification of the protein was achieved by matching the uninterpreted product ion spectrum against protein sequence databases with varying degrees of annotation, coupled with a scoring scheme weighted for the relative abundances of the experimentally observed product ions and the frequency of fragmentations occurring at preferential sites. The protein was identified to be an acetylated yeast heat shock protein, HS12_Yeast (11.6 kDa), with the initiating methionine residue removed. This constitutes the first example of the identification of an a priori unknown protein that is not present in an annotated protein database using a “top-down” approach with a quadrupole ion trap. This example illustrates the utility of relatively low cost instrumentation with modest mass analysis characteristics for the identification of modified proteins without recourse to enzymatic digestion. It also illustrates how experimental data can be used interactively with protein databases when the modified protein of interest is not initially present in the database. Mass spectrometry has become a very attractive tool in proteomics studies due to some of its attractive performance characteristics such as specificity, speed, and sensitivity.1-4 For pure proteins and proteins in very simple mixtures, identification is often accomplished by peptide mass fingerprinting,5-9 following * Correcponding author. Phone: (765) 494-5270. Fax: (765) 494-0239. E-mail: [email protected]. (1) Aebersold, R.; Goodlett, D. R. Chem. Rev. 2001, 101, 269-295. (2) Yates, J. R., III. J. Mass. Spectrom. 1998, 33, 1-19. (3) Peng, J.; Gygi, S. P. J. Mass Spectrom. 2001, 36, 1083-1091. (4) Pandey, A.; Mann, M. Nature 2000, 405, 837-846. (5) Henzel, W. J.; Billeci, T. M.; Stults, J. T.; Wong, S. C.; Grimley, C.; Watanabe, C. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 5011-5015.

720 Analytical Chemistry, Vol. 76, No. 3, February 1, 2004

a 1D or 2D gel electrophoresis and proteolytic digestion of individual protein spots. In the case of proteins present in more complex mixtures, tandem mass spectrometry (MS/MS) of enzymatically or chemically derived individual peptides followed by protein database analysis of the product ion spectra is employed.10-13 By far, most protein identifications accomplished via mass spectrometry have employed peptides formed from protein digestion using one of the two overall approaches just mentioned. These approaches have been termed as “bottom-up” to distinguish them from tandem mass spectrometry of whole protein ions, which has been termed the “top-down” approach.14 One advantage that the top-down approach has over bottom-up approaches in direct application to protein mixtures is that the former involves the fragmentation of whole protein ions in the gas phase without prior recourse to enzymatic digestion, which further complicates the already complex mixture. The tandem mass spectrometry of whole protein ions, however, is more challenging than the tandem mass spectrometry of peptide ions, at least for most forms of instrumentation.15 Nevertheless, technologies to execute the top-down approach are maturing. Most work has been done using high magnetic field strength Fourier transform mass spectrometry.16-19 However, recent examples of top-down approaches using quadrupole/time-of-flight,20 quadrupole (6) James, P.; Quadroni, M.; Carafoli, E.; Gonnet, G. Biochem. Biophys. Res. Commun. 1993, 195, 58-64. (7) Mann, M.; Hojrup, P.; Roepstorff, P. Biol. Mass Spectrom. 1993, 22, 338345. (8) Pappin, D. J. C.; Hojrup, P.; Bleasby, A. J. J. Curr. Biol. 1993, 3, 327-332. (9) Yates, J. R., III; Speicher, S.; Griffin, P. R.; Hunkapillar, T. Anal. Biochem. 1993, 214, 397-408. (10) Mann, M.; Wilm, M. Anal. Chem. 1994, 66, 4390-4399. (11) Eng, J. K.; McCormack, A. L.; Yates, J. R., III. J. Am. Soc. Mass Spectrom. 1994, 5, 976-989. (12) Yates, J. R., III; Eng, J. K.; McCormack, A. L. Anal. Chem. 1995, 67, 32023210. (13) Link, A. J.; Eng, J.; Schieltz, D. M., Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R., III. Nat. Biotechnol. 1999, 17, 676-682. (14) Kelleher. N. L.; Lin, H. Y.; Valaskovic, G. A.; Aaserud. D. J.; Fridriksson, E. K.; McLafferty, F. W. J. Am. Chem. Soc. 1999, 121, 806-812. (15) Reid, G. E.; McLuckey, S. A. J. Mass Spectrom. 2002, 37, 663-675. (16) Zubarev, R. A.; Kelleher, N. L., McLafferty, F. W. J. Am. Chem. Soc. 1998, 120, 3265-3266. (17) Zubarev, R. A.; Kruger, N. A.; Fridriksson, E. K.; Lewis, M. A.; Horn, D. M.; Carpenter, B. K.; McLafferty, F. W. J. Am. Chem. Soc. 1999, 121, 28572862. (18) Ge, Y.; Lawhorn, B. G.; ElNagger, M.; Strauss, E.; Park, J. H.; Begley, T. P.; McLafferty, F. W. J. Am. Chem. Soc. 2002, 124, 672-678. (19) Meng, F.; Cargile, B. J.; Miller, L. M.; Forbes, A. J.; Johnson, J. R.; Kelleher, N. L. Nat. Biotechnol. 2001, 19, 952-957. (20) Nemeth-Cawley, J. F.; Rouse, J. C. J. Mass Spectrom. 2002, 37, 270-282. 10.1021/ac034900k CCC: $27.50

© 2004 American Chemical Society Published on Web 01/03/2004

ion trap,21-23 and time-of-flight/time-of-flight instruments24 have been described. Our laboratory has been engaged in the development of quadrupole ion trap technology for the analysis of ions of relatively high mass molecules, such as whole proteins, formed by electrospray ionization. An enabling aspect of this work is the use of ion/ion proton-transfer chemistry25 for charge-state manipulation. Ion/ion proton-transfer reactions are useful for dealing with complications arising from the multiple charging phenomenon associated with electrospray ionization of biomolecules by reducing charge states to simplify spectral interpretation.26-28 Ion/ion reactions can also be used to concentrate the signals from several charge states into a single charge state for subsequent tandem mass spectrometry as part of a technique referred to as “ion parking”.29 The application of the overall ion trap methodology employing ion/ion reactions to a mixture of proteins derived from Escherichia coli was recently described30 in which several proteins present in the unannotated database were identified in the mixture. In this study, we illustrate the identification of a modified protein that is not present in a partially annotated protein database using top-down protein analysis in the quadrupole ion trap. Specifically, we describe the identification and characterization of an a priori unknown modified protein present in a HPLC fraction of a whole cell lysate of a yeast (Saccharomyces cerevisiae) strain containing ∼19 proteins. We illustrate the experimental and database search strategies that enable the identification of modified proteins using a quadrupole ion trap adapted with ion/ion reaction capabilities for charge-state manipulation. EXPERIMENTAL SECTION Reagents. Acetic acid and HPLC grade acetonitrile were obtained from Mallinckrodt (Paris, KY). Trifluoroacetic acid (TFA) was purchased from Pierce (Rockford, IL). Growth and Lysis of S. cerevisiae. S. cerevisiae strain S288C was purchased from American Type Culture Collection (Rockville, MD) and reactivated on agar plates at 30 °C for 24 h under sterile conditions. The synthetic media, containing 1.7 g of yeast nitrogen base, 20 g of dextrose, 5 g of ammonium sulfate, and 17 g of agar (DOBA from Q‚BIOgene Inc., Carlsbad, CA) per liter, was employed to prepare the agar plates and grow yeast. A single colony was extracted out from this plate and suspended in 500 mL of growth media (the same ingredients without agar, DOB from Q‚BIOgene Inc.) in a 1-L culture flask. Yeast cells were grown to midlog phase with optical density of 1.0 at 600 nm in an (21) Stephenson, J. L., Jr.; McLuckey, S. A. Anal. Chem. 1998, 70, 3533-3544. (22) VerBerkmoes, N. C.; Strader, M. B.; Smiley, R. D.; Howell, E. E.; Hurst, G. B.; Hettich, R. L.; Stephenson, J. L., Jr. Anal. Biochem. 2002, 305, 68-81. (23) Schey, K. L.; Cook, L. A.; Hildebrandt, J. D. Int. J. Mass Spectrom. 2001, 212, 377-388. (24) Lin, M.; Campbell, J. M.; Mueller, D. R.; Wirth, U. Rapid Commun. Mass Spectrom. 2003, 17, 1809-1814. (25) Stephenson, J. L., Jr.; McLuckey, S. A.J. Am. Chem. Soc. 1996, 118, 73907397. (26) McLuckey, S. A.; Stephenson, J. L., Jr.; Asano, K. G. Anal. Chem. 1998, 70, 1198-1202. (27) Stephenson, J. L., Jr.; McLuckey, S. A. J. Am. Soc. Mass Spectrom. 1998, 9, 585-596. (28) Schaaff, T. G.; Cargile, B. J.; Stephenson, J. L., Jr.; McLuckey, S. A. Anal. Chem. 2000, 72, 899-907. (29) McLuckey, S. A.; Reid, G. E.; Wells, J. M. Anal. Chem. 2002, 74, 336-346. (30) Reid, G. E.; Shang, H.; Hogan, J. M.; Lee, G. U.; McLuckey, S. A. J. Am. Chem. Soc. 2002, 124, 7353-7362.

aerobic environment at 30 °C. The yeast cells were then harvested by centrifugation at 4000g for 15 min at 4 °C and resuspended in 15 mL of ice-cold water. This procedure was repeated three times. The final yeast cell pellet (∼2 g; wet weight) was resuspended in 10 mL of lysis buffer (200 mM Tris, pH 8 from Sigma (St. Louis, MO), 150 mM ammonium sulfate from Mallinckrodt (Paris, KY), 10% (v/v) glycerol and 1 mM EDTA from Aldrich (Milwaukee, WI), and protease inhibitor cocktail from Calbiochem (San Diego, CA)), and cells were lysed by French press (18 000 psi). The cellular debris, including unlysed cells, was cleared by centrifugation at 500g for 15 min. The supernatant containing soluble yeast proteins was pipetted out. Before fractionation of an aliquot of supernatant on RP-HPLC, the aliquot was again centrifuged at 10 000 rpm for 10 min. The protein concentration of this S. cerevisiae whole cell lysate was estimated to be 4-6 mg/mL by Bio-Rad Bradford assay using bovine serum albumin as a standard. Fractionation of Yeast Whole Cell Lysate by RP-HPLC. About 800 µg (200 µL/10 mL total) of soluble proteins from the yeast whole cell lysate was fractionated by reversed-phase HPLC on a Hewlett-Packard (Palo Alto, CA) model 1090 HPLC, using a Poros (Applied Biosystems, Foster City, CA) R1/10 100 mm × 2.1 mm i.d. column operated at 0.5 mL/min. A linear 12-min gradient from 0 to 100% buffer B was used, where buffer A was 0.1% aqueous TFA and buffer B was 60% acetonitrile/40% H2O containing 0.09% aqueous TFA. The fractions were collected at 20-s intervals while the column was maintained at 40 °C. The UV absorbance was monitored at 215 nm. The collected fractions were lyophilized to dryness and then dissolved in 200 µL of aqueous 1% aqueous acetic acid for introduction to the mass spectrometer. Mass Spectrometry. Experiments were performed using a Hitachi Instruments Inc. (San Jose, CA) model M-8000 quadrupole ion trap mass spectrometer with a home-built nanoelectrospray ion source. The instrument has been modified for anion introduction through a hole in the ring electrode, as described previously.31 Anions derived from perfluoro-1,3-dimethylcyclohexane (PDCH) via atmospheric sampling glow discharge ionization32 were used in this work. Protein ions were accumulated for 1 s followed by PDCH anion injection time of a few milliseconds. The ions were allowed to undergo proton-transfer reactions for 350 ms while a single-frequency resonance excitation voltage was applied to ion park29 a particular m/z ion of interest, which in this case was the [M + 9H]9+ ion of the unknown protein. Following the ion/ion reaction period, all low-m/z ions, including residual PDCH anions, were ejected from the trap by increasing the amplitude of the radio frequency applied to the ring electrode of the ion trap during an ion isolation step. Two ion isolation steps of 20 ms were carried out using the Hitachi instrument’s filtered noise fields. This was followed by an ion activation step involving application of a singlefrequency (68.2 kHz) resonance excitation voltage of 1.3 Vpeak-to-peak for 300 ms, corresponding to the z-dimensional secular frequency of the ion of interest to the end cap electrodes, via an external waveform generator (model 33120A; Agilent, Palo Alto, CA). A final ion/ion reaction step with PDCH anions was used to reduce the multiply charged product ion population to predominantly their (31) Reid, G. E.; Wells, J. M.; Badman, E. R.; McLuckey, S. A. Int. J. Mass Spectrom. 2003, 222, 243-258. (32) McLuckey, S. A.; Glish, G. L.; Asano, K. G.; Grant, B. C. Anal. Chem. 1988, 60, 2220-2228.

Analytical Chemistry, Vol. 76, No. 3, February 1, 2004

721

Figure 1. Pre- and post-ion/ion reaction mass spectra of the 20-s fraction starting at 8.00 min from RP-HPLC of the S. cerevisiae whole cell lysate soluble proteins. The pre-ion/ion mass spectrum (a) was acquired using a resonance ejection frequency of 305 kHz. The postion/ion mass spectrum (b) was acquired at 35 kHz following an ion/ ion reaction period of 100 ms with PDCH anions. The regions over which doubly and triply charged ions can also contribute to the spectrum are also indicated.

singly charged forms, thereby facilitating fragment mass assignment. Finally, a product ion spectrum was acquired by resonance ejection33 after residual PDCH anions were ejected by increasing the amplitude of the radio frequency applied to the ring electrode of the ion trap. Calibration of the post-ion/ion product ion scans was accomplished using the singly, doubly, and triply charged ions of intact bovine cytochrome c formed by ion/ion reactions sans CID. RESULTS AND DISCUSSION Isolation and Tandem Mass Spectrometry of the Unknown Protein from a Whole Cell Lysate Fraction. The electrospray mass spectrum derived from the 20-s LC fraction of the yeast protein lysate beginning at a retention time of 8.00 min is shown in Figure 1a. Only one potential charge-state distribution is apparent in this spectrum, and this corresponds to the 8+-10+ ions of a molecule of mass 11.6 kDa. Figure 1b shows the electrospray mass spectrum after the ions reflected in Figure 1a were subjected to reactions with the [M - F]- and [M - CF3]anions derived from glow discharge ionization of PDCH to reduce charge states to 3+-1+. Examination of this spectrum reveals signals for roughly 19 proteins over the range of 6-12 kDa using a criterion of >5% of the most abundant ion signal. The masses associated with the most abundant of these are indicated in the figure. The m/z ) 1291 Da ion, corresponding to the [M + 9H]9+ species of the 11.6 kDa protein, was ion parked29 by subjecting the initial ion population to ion/ion reactions for 350 ms and then isolated twice to eliminate ions of higher and lower m/z ratio (Figure 2a. The ion parking process served to both transfer most of the [M + 10H]10+ ions to the 9+ charge state and simultaneously reduce the charge states of all other ions, such that most of the (33) Kaiser R. E., Jr.; Cooks, R. G.; Stafford, G. C., Jr.; Syka, J. E. P.; Hemberger, P. H. Int. J. Mass Spectrom. Ion Processes 1991, 106, 79-115.

722 Analytical Chemistry, Vol. 76, No. 3, February 1, 2004

Figure 2. Ion parked and isolated [M + 9H]9+, m/z ) 1291, of unknown protein (a) in the 20-s fraction (starting from 8.00 min) Preion/ion reaction product ion spectrum (b) upon CID at 68.2 kHz frequency resonance excitation voltage of 1.3 Vpeak-to-peak for 300 ms, corresponding to the center of mass of m/z ) 1291 Da. Both spectra were acquired at 305 kHz.

ion signal in Figure 1a was moved to significantly lower m/z values than that of the [M + 9H]9+ ion. This greatly simplifies the isolation of a relatively pure ion population for subsequent ion activation. Further purification is possible via the use of a second ion parking period whereby, for example, in this case, the [M + 8H]8+ ion could be parked after isolation of the ions corresponding in m/z to the [M + 9H]9+ ion were isolated. The second ion parking29 step can serve to disperse ions of different mass and charge but similar m/z ratio in the ion population containing the [M + 9H]9+ ion. This was demonstrated in the previously reported E. coli study.30 This experiment needed only a single ion parking step as only one protein charge-state distribution was observed in the post-ion/ion spectrum of the isolated 1291 m/z peak (data not shown). Collisional activation of the isolated ion population was then performed using a 68 200 Hz frequency with an excitation voltage of 1.3 Vpeak-to-peak for 300 ms. The resulting product ion spectrum is shown in Figure 2b. The resolving power of the ion trap used in this work is insufficient to determine unambiguously the charge states of product ions greater in mass than ∼2000 Da. Furthermore, overlaps in m/z arising from product ions of different mass and charge complicate product ion mass assignment. Therefore, the product ions derived from fragmentation of large multiply charged polypeptide ions, like those in Figure 2b, can be subjected to ion/ ion proton-transfer reactions to simplify the determination of product ion masses. Figure 3 shows the product ion spectrum of the [M + 9H]9+ ion acquired after the product ions reflected in Figure 2b were subjected to ion/ion proton-transfer reactions. In this spectrum, the product ions are reduced largely to the +1 and +2 charge states. All of the ions of m/z greater than that of the residual doubly charged parent ions must be singly charged. Furthermore, doubly and triply charged ions arising from these high-mass fragments can be readily identified on the basis of m/z ratio and abundance relative to the residual doubly and triply charged parent ions. The relative abundances of the product ions in the post-ion/ion product ion spectrum are expected to provide a more accurate measure of the true product ion relative

Figure 3. Post-ion/ion reaction MS/MS spectrum of the [M + 9H]9+ ion, m/z ) 1291 Da, of the unknown protein following ion/ion reaction period of 100 ms with anions derived from glow discharge ionization of PDCH vapors.

abundances than those of the pre-ion/ion product ion spectrum. Detector discrimination for ions of different charge states, which is not well characterized, can lead to misleading assessments of product ion abundances. Therefore, the post-ion/ion product ion spectrum gives more reliable mass and abundance assignments for the product ions than does the pre-ion/ion product ion spectrum. The masses and abundances associated with the labeled peaks were used for identification of the protein (see below). Product ion masses and abundances were manually chosen from the post-ion/ion MS/MS spectra according to the following criteria. First, no peaks corresponding to doubly and triply charged fragment or unfragmented protein ions were used in the database search. Such ions are generally readily recognized by their m/z ratios and abundances relative to those of singly charged ions that fall in the m/z range between the residual doubly and singly charged parent ions. Second, product ion peaks were distinguished from noise peaks by examination of the peak shape and abundance level, with the criterion that the abundance of the product ion peak be >10% of the most abundant peak. Third, no peaks from below m/z 4700 were chosen due to the increased spectral congestion arising from overlap of singly and multiply charged ions. (Note that this congestion could be reduced by the use of longer ion/ion reaction times but this was not necessary to obtain a sufficient number of product ions for a database search.) Finally, mass and abundance values for the selected product were assigned from the location of the peak apex. Protein Identification by Database Searching of Uninterpreted Whole Protein MS/MS Spectra. Unknown protein identification via a top-down approach involves searching the uninterpreted product ion mass spectra of whole proteins against in silico fragmentation of the proteins contained in either a protein database or a translated genome database. Typically, an experimentally determined protein precursor ion mass with a userspecified mass tolerance is used to select probable protein matches based on protein mass. The size of the mass tolerance can be changed to account for possible posttranslational modifications, or parent mass can be omitted as a constraint if one wishes to search the entire database. The matching candidate proteins are then fragmented in silico into either b- and y-type ions, or c- and

Table 1. Fragment Ion Mass List and Abundances Used in the Database Search fragment mass (Da)

abundance (arb units)

fragment mass (Da)

abundance (arb units)

4721.9 5084.4 5131.8 5174.8 5441.2 6034.6 6165.3 6401.5 6518.7 6877.3

17.3 3.9 3.6 3.4 2.8 2.8 2.5 3.3 3 16.2

7086.1 7565.4 7882.7 7946.8 9599.6 10948.3 11014.5 11129.5 11359.1 11458.8

3.9 3.8 6.2 5.4 6.2 5.1 6.2 7.9 4.9 10.5

z•-ions.19,30 The experimental and theoretical fragment ion masses are compared, and matched fragments are used to assign a score to the candidate protein match. The initial database search of the uninterpreted post-ion/ion product ion spectrum masses from the unknown yeast protein was done using software developed in-house as described previously.30 All database searches involved using a precursor mass of 11 606 Da, a precursor mass tolerance of (50 Da, and a fragment mass tolerance of (10 Da. Only theoretical b- and y-type ions were searched since they are most commonly observed in ion trap CID of whole protein positive ions. The software-scoring algorithm, described previously, uses fragment ion abundances and gives added weight to preferred cleavages in the protein identification.30 The product masses used in the database search along with ion abundances are shown in Table 1. The database searched was a modified version of the SwissProt database (v. 41.0), an annotated database, where only yeast (S. cerevisiae) proteins (4915 proteins) were used and each protein sequence was modified, in-house, for N-terminal methionine removal and signal and propeptide removal (9285 total protein forms). The database search resulted in 14 candidate protein matches. The highest scored protein was HS12•Yeast (accession number P22943), with the N-terminal methionine removed, which matched 10 of 20 fragment ions with a score of 195.2. The protein had nine unique fragment ion Analytical Chemistry, Vol. 76, No. 3, February 1, 2004

723

Table 2. Probability and Expectation Values from the Database Search of SwissProt (v. 41.0) Yeast Proteins

name

McLuckey score

HS12•Yeasta YCU0•Yeastb

195.2 115.81

a

Poisson (Meng) probability expectation 0.0770 0.0303

1.0787 0.4242

Poisson (Sadygov) probability expectation 0.1003 0.0370

1.4042 0.518

hypergeometric probability expectation 0.1048 0.0345

1.4672 0.483

HS•Yeast - N-terminal methionine. b YCU0•Yeast - N-terminal methionine.

Table 3. Probability and Expectation Values from the Database Search of the Saccharomyces Genome Database

name

McLuckey score

ProSight PTM score

SW-HS12•Yeasta GP-CAA96925•1 PIR-s69299 GP-CAA62766•1b

200.757 150.294 73.760 12.784

2.4714 3.7754 0.2028 0.2632

a

Poisson (Meng) probability expectation 0.0504 0.0770 0.0041 0.0054

2.4714 3.7754 0.2028 0.2632

Poisson (Sadygov) probability expectation 0.0668 0.0775 0.0070 0.0041

3.2732 3.7975 0.3455 0.2009

Hypergeometric probability expectation 0.0981 0.1115 0.0088 0.0052

4.8069 5.4635 0.4312 0.2548

SW-HS12•Yeast - N-terminal methionine. b GP-CAA62766•1 - N-terminal methionine.

matches as one experimental fragment ion matched two possible theoretical fragments. The second best match was YCU0•Yeast (accession number P25630), with the N-terminal methionine removed, matching 12 of 20 fragment ions (all unique) and a score of 115.81. Although the scoring algorithm identifies HS12•Yeast as the unknown protein, only 10 of the possible 20 fragment ions are identified. The identified fragment ions are all y-type ions except for one b-ion corresponding to the duplicate fragment ion match. Furthermore, the theoretical precursor protein mass also differs from the experimental mass by 43 Da. This mass difference implies the protein is either not in the database or is posttranslationally modified in a way not accounted for by the software. Probability statistics utilizing Poisson distributions published by Meng et al.19 and Sadygov and Yates34 and the hypergeometric probability model published by Sadygov and Yates34 were used to assess the likelihood that this protein assignment is a false positive identification. The Poisson distribution models published by both Meng and Sadygov provide similar probabilities, although they use different values as inputs in the Poisson distribution. These probability distributions were used to evaluate the search results in the absence of a probability-based approach that takes into account preferred cleavages. The probability results for each probability distribution are shown in Table 2. An expectation value can be determined from the probability results by multiplying the probability of each protein candidate by the total number of protein candidates within the parent ion mass window. For peptides, an expectation value of