In Proteomics, while doing analysis on protein pilot after running my sample on mass spec; I got plenty of protein IDs which are having zero peptide identification to the library. I don't understand how software can report particular ID when there is no match?
Joel,
it will almost certainly be a case that there are matching peptides, but once the FDR is set, presumable 5% looking at your results, the matches are no longer statistically valid. So the protein *may* be present, but the scores for the matching peptides are NOT significant, hence the 0 matching peptides returned.
Joel,
it will almost certainly be a case that there are matching peptides, but once the FDR is set, presumable 5% looking at your results, the matches are no longer statistically valid. So the protein *may* be present, but the scores for the matching peptides are NOT significant, hence the 0 matching peptides returned.
A 95% confident peptide corresponds to a contribution of 1.33 (-log confidence). The proteins hve a protein score of 0.53 and 0.06, which corresponds to individual peptide confidences of 10e-0.53 or even less - definitely below 95%, which is the threshold for column H. It simply means there are no confident peptide hits for these proteins at all.
Question is- did you perform a False Discovery Rate estimation, and use the cutoff resulting? My question would be why you are reporting such low-scoring proteins in the first place. Please look up the FDR-related sections in the ProteinPilot documentation, and the corresponding publication by Sean Seymour et al. on how PP calculates local FDR. That should help.
- Badges
- Science topic
- Similar topics
- Chemistry
- General Biochemistry
- Proteomics