The ProSightPD application performs label-free quantitation (LFQ) using the sliding window deconvolution algorithm.

In an LFQ experiment, the proteoforms are quantified across their elution profile. In ProSightPD, these quantitation results are called Feature Groups. Feature Groups are generated nodes using the sliding window algorithm in conjunction with the Xtract (for high resolution, isotopically-resolved data) or kDecon (for low or medium resolution, isotopically unresolved data) algorithms for deconvolution.

The sliding window algorithm averages spectra over a succession of windows in retention time, deconvolves each average spectrum, and then merges similar masses from consecutive deconvolutions to form feature groups.

The following figure illustrates the concept of sliding windows.

Figure Sliding window overview
Sliding window overview

NOTE

FWHM refers to full-width half-max.

The sliding window algorithm benefits top-down LFQ in several ways:

  • Reduces the number of false positives due to noisy data
  • Improves sensitivity using signal averaging
  • Sensitive to low abundance co-eluting species identifies co-eluting species
  • Accurately defines elution profiles for quantitation

The ProSightPD Hi Res. Feature Detector node and the ProSightPD Med. Res. Feature Detector node contain the parameters controlling the sliding window algorithm.

After the Feature Detector node determines the feature groups for all data files, the node maps the feature groups to individual PrSMs and connects feature groups to quantitation traces. The node groups the features that are similar in mass and retention time between files. A group of feature groups across all files is a consensus feature group.

The Feature Mapper node maps the consensus feature groups to proteoforms based on a mass and retention time threshold.

The ProSightPD Quantifier node produces abundance ratios and statistics using the measured abundances and study factors you applied.