Peakfinding

Chromatographic peak-finding is performed for each input PSM.

For each charge state between the global minimum and maximum observed, the peptide’s monoisotopic m/z is calculated and the mass-spectral peaks with that m/z are retrieved from the peak index. These mass-spectral peaks are filtered by a large mass tolerance (default 20 ppm).

Starting from the PSM’s precursor MS1 scan, the list of indexed mass-spectral peaks is iterated through in order of scan number (i.e., retention time), stopping when a gap is detected in retention time space (by default, 1 “missed” scan is allowed). The same method is applied in reverse order, to find mass spectral peaks before the PSM in retention-time space.

This initial peak is further refined by a narrow mass tolerance (10 ppm by default) and by expected isotopic distribution for that peptide sequence and charge.

Each of these “raw” chromatographic peaks are examined for a discrimination factor of 0.6 or higher to detect an “intensity valley” that may correspond to two closely-eluting peptides, and the peak is cut if one of these valleys is found.

These “raw” peaks are merged together if they have duplicate apex mass-spectral peaks (for example, two PSMs from the same peptide that found the same chromatographic peak, or two co-eluting peptides that were not distinguishable).