Non specific cleavage ("No Enzyme", "Non specific") searches

There are many applications where a non-specific cleavage search, often referred to as a no-enzyme search, is useful. Neoantigens presented by the major histocompatibility complex (MHC-I) require either a no-enzyme search or complete de novo approach for peptide identification. The non-specific search option in MetaMorpheus can be found in the "Search Parameters" menu inside the pulldown for "protease".

Recommendations when using the non-specific search option:

• Use the Non-Specific Search Mode. It an order of magnitude faster than the classic search.
• Increase the number of database partitions to something obnoxious, say 20, to prevent RAM issues.
• set the Min Peptide Len and Max Peptide Len: reasonable limits will enable your search to finish in a reasonable amount of time. MHC-I peptides are generally between 8 and 16 amino acids long. There will of course be outliers, but the bulk of results can be found in this window.
• Set the Search Parameters/Max Missed Cleavages to a value one less that the Max Peptide Len. For example, if the Max Peptide Len is 16, then set the Max Missed Cleavages to 15.
• calibrating raw files for no-enzyme searches: This can take "forever" if work-arounds are not deployed. The following steps yield good results. Do a full search with slightly wide parent tolerance for a single peptide length (e.g. 9AA long). Take the list of assession numbers given in the unique peptides list and create a second smaller database containing only those accession numbers. Calibrate the raw files using the database subset and the fixed peptide length (e.g. 9AA long). After calibration, you can go back and search with the full database and using a reasonable range of peptide lengths.
• an excellent qualitative check of the results is a plot of the actual peptide retention time versus the time predicted by SSRCalc.