Hi,
one of the problems in targeted proteomics is the question: is the
selected protein MS-friendly at all? Meaning if I select a peptide
which is not seen in ESI-MS, I will never get a result back. I am
currently using the ESPPredictor from the Steven Carr lab
(
www.genepattern.org) and would like to know if anyone else has used
it. For me it looks very convincing and I am going into more analysis
of the predicted peptides. Would it be possible for TIQAM (after
talking with Steven Carr and Vince Fusaro) to integrate such a tool?
Manfred