I have data from LC/MS-MS, that gave 55 results, including some contaminations. The main objetive is to try to identify proteins with 56-60 kDa. What are the coverage, #peptides, # PSMs, # Unique Peptides, Score Sequest HT: Sequest HT, # Peptides (by Search Engine): Sequest HT, etc, numbers or scores that I should have for criteria?
Hi Ana
When analysing LC-MS data, the most important thing to keep in mind is your underlying research question. MS/MS data will give a lot of unnecessary information which could be quite intimidating if you keep focussing on those.
Firstly, you need to make sure that the database that you are querying for peptide assignment is properly annotated. The values associated with all the parameters that you mentioned rely on the resolution and peptide detection ability of the Mass spectrometry machine. I am guessing you might have run standards on the machine before analysing your samples. You can take those as reference for your study.
If you have a proper annotated database, you can keep the coverage around 50-70%, unique peptides 3-5 and SEQUEST HT score greater than 100 as a filter. To increase the confidence of the detection, you can manually check the peptides that were matched to your protein to have appropriate coverage of b- and y- ions.
Disclaimer: This is my personal opinion and there is no universal threshold that is set for these parameters. It changes with the factors that I discussed.
The minimum criteria I set are: high FDR confidence, no contaminations and at least 1 unique peptide evidence. You can be more stringent by adding other criteria.
For 56-60kDa proteins, you can also use amino acid length cutoff (generally 1kDa roughly equals to 9-10 amino acids so you are looking for ~500-600 aa long proteins).
To elaborate on abbreviations, the coverage means % of protein sequence supported by detected tryptic peptides; #peptides is how many per protein have been detected; # PSMs is the number of peptide-spectrum matches, which can be different from the # peptides as one spectrum can be matched to several peptides, or a peptide could have been selected for MS/MS more than once thus generating several matching spectra; # Unique Peptides is how many of the detected peptides match to only one protein in the database. As others mentioned, with tryptic protein digestion, one is expecting to detect several peptides matched to each protein sequence, detecting single peptide per protein means really low confidence of identification for a large protein, even with good MS/MS.
- Badges
- Science method