Abstract
Shotgun proteomics uses liquid chromatography–tandem mass spectrometry to identify proteins in complex biological samples. We describe an algorithm, called Percolator, for improving the rate of confident peptide identifications from a collection of tandem mass spectra. Percolator uses semi-supervised machine learning to discriminate between correct and decoy spectrum identifications, correctly assigning peptides to 17% more spectra from a tryptic Saccharomyces cerevisiae dataset, and up to 77% more spectra from non-tryptic digests, relative to a fully supervised approach.
This is a preview of subscription content
Access options
Subscribe to Journal
Get full journal access for 1 year
$119.00
only $9.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Buy article
Get time limited or full article access on ReadCube.
$32.00
All prices are NET prices.
References
Eng, J.K., McCormack, A.L. & Yates, J.R. III. J. Am. Soc. Mass Spectrom. 5, 976–989 (1994).
Perkins, D.N., Pappin, D.J.C., Creasy, D.M. & Cottrell, J.S. Electrophoresis 20, 3551–3567 (1999).
MacCoss, M.J., Wu, C.C. & Yates, J.R. III. Anal. Chem. 74, 5593–5599 (2002).
Keller, A., Nezvizhskii, A.I., Kolker, E. & Aebersold, R. Anal. Chem. 74, 5383–5392 (2002).
Moore, R.E., Young, M.K. & Lee, T.D. J. Am. Soc. Mass Spectrom. 13, 378–386 (2002).
Peng, J., Elias, J.E., Thoreen, C.C., Licklider, L.J. & Gygi, S.P. J. Proteome Res. 2, 43–50 (2003).
Anderson, D.C., Li, W., Payan, D.G. & Noble, W.S. J. Proteome Res. 2, 137–146 (2003).
Boser, B.E., Guyon, I.M. & Vapnik, V.N. A training algorithm for optimal margin classifiers. in 5th Annual ACM Workshop on COLT (ed. Haussler, D.) 144–152 (ACM Press, Pittsburgh, Pennsylvania, USA, 1992).
Storey, J.D. & Tibshirani, R. Proc. Natl. Acad. Sci. USA 100, 9440–9445 (2003).
Tabb, D.L., McDonald, W.H. & Yates, J.R. III. J. Proteome Res. 1, 21–26 (2002).
Washburn, M.P., Wolters, D. & Yates, J.R. III. Nat. Biotechnol. 19, 242–247 (2001).
Acknowledgements
This work was funded by US National Institutes of Health grants P41 RR011823 and R01 EB007057.
Author information
Authors and Affiliations
Contributions
M.J.M. came up with the initial idea to use decoy PSMs as negative examples. L.K. and W.S.N. came up with the idea to use a support vector machine using semi-supervised learning. L.K. implemented Percolator and performed computational experiments. J.W. provided machine learning expertise. J.D.C. performed initial proof-of-concept experiment and provided mass spectrometry expertise. W.S.N., L.K. and M.J.M. wrote the article.
Corresponding author
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–4, Supplementary Tables 1 and 2, Supplementary Methods, Supplementary Data (PDF 1393 kb)
Rights and permissions
About this article
Cite this article
Käll, L., Canterbury, J., Weston, J. et al. Semi-supervised learning for peptide identification from shotgun proteomics datasets. Nat Methods 4, 923–925 (2007). https://doi.org/10.1038/nmeth1113
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/nmeth1113
Further reading
-
Integrated proteogenomic characterization of urothelial carcinoma of the bladder
Journal of Hematology & Oncology (2022)
-
Integrated proteomic analysis of low-grade gliomas reveals contributions of 1p-19q co-deletion to oligodendroglioma
Acta Neuropathologica Communications (2022)
-
GATD3A, a mitochondrial deglycase with evolutionary origins from gammaproteobacteria, restricts the formation of advanced glycation end products
BMC Biology (2022)
-
TIDD: tool-independent and data-dependent machine learning for peptide identification
BMC Bioinformatics (2022)
-
Mutations in Hcfc1 and Ronin result in an inborn error of cobalamin metabolism and ribosomopathy
Nature Communications (2022)
