Proteomics Fundamentals

How to determine and optimize the confidence in protein identifications?
While proteins are identified routinely by many laboratories there is no consistent view on what amount/quality of data is sufficient to call a protein identified. In fact, it is not even clear what it means to identify a protein[1]. The lack of a standard in the field mitigates exchange of results. We require for our own work on chromatin and thus develop a method that gives us a reliable judgement of confidence.

How to determine the stoichiometry of proteins in a complex mixture?
Proteins can be quantitated using mass spectrometric data by a variety of different approaches. This is true, however, only as long “same and same” are compared, i.e. the same peptide/protein from different samples. We are interested in the stoichiometry of proteins in complex mixtures which requires to compare “apples and oranges”. We made the first step for this some time ago with the introduction of the Protein Abundance Index (PAI)[2] and maintain our interest to further improve the method.

How to determine the completeness of an analysis?
Our first analysis of the spliceosome identified 42 proteins, our second analysis 312 proteins[2]. But how many proteins are there in total? For small protein complexes the answer will be found through functional studies, for large protein complexes this is not possible. We aim to estimate the yield of a proteomics experiment based on bioinformatics.

[1] Rappsilber J, Mann M.
What does it mean to identify a protein in proteomics?
Trends Biochem Sci. 2002 Feb;27(2):74-8. Review.

[2] Rappsilber J, Ryder U, Lamond AI, Mann M.
Large-scale proteomic analysis of the human spliceosome.
Genome Res. 2002 Aug;12(8):1231-45.