Analysis of the Saccharomyces cerevisiae proteome with PeptideAtlas

Table 4 Expected errors

Case	Expected errors P_limit = 0.9	Expected errors P_limit = 0.95	Expected errors P_limit = 0.99
MS/MS P_i ≥ P_limit	0.00915 (9%)	0.00517 (5%)	0.00137 (1%)
MS/MS P_i ≥ P_limit, N_{peptide observed} > 1	0.00884 (9%)	0.00506 (5%)	0.00136 (1%)
Consensus peptide best P_i, P_i ≥ P_limit	0.01027 (10%)	0.00510 (5%)	0.00120 (1%)
Consensus peptide best P_i, P_i ≥ P_limit, N_{peptide observed} >1	0.00272 (3%)	0.00215 (2%)	0.00078 (1%)

Expected errors in the Saccharomyces PeptideAtlas are calculated for four cases and lower probability limits (P_limit = 0.9, 0.95, 0.99): the assigned probability P of each MS/MS is used for all P_i ≥ P_limit; the assigned probability P of each MS/MS is used for all P_i ≥ P_limit where the associated peptide has been seen in the S. cerevisiae PeptideAtlas more than once'; the best probability for each unique peptide sequence is used for all P_i ≥ P_limit; the best probability for each unique peptide sequence is used for all P_i ≥ P_limit when the peptide has been seen in the S. cerevisiae PeptideAtlas more than once.

ISSN: 1474-760X