Protein Coverage Explorer

Coverage

Each peptide line is parsed with peptacular's ProForma parser (peptacular.parse) and its unmodified residue sequence is searched as an exact substring of the protein (optionally with I and L treated as the same residue, since they have identical mass). Lines that are not valid ProForma fall back to removing bracketed text and non-letters. Every match adds the peptide's PSM count to each residue it covers, so the depth of a residue is the summed PSM count of all peptides over it. A peptide that matches several places counts at each of them and is flagged as ambiguous. Sequence coverage is the percentage of residues with depth above zero. Binary coverage sets every covered residue to depth 1.

Property tracks

Computed with peptacular.calc_window_property(): the mean scale value over a sliding window of w residues, drawn at the window's centre residue. Residues missing from a scale count as 0. Colours are scaled from the lowest (blue) to the highest (red) value within this protein, so they compare regions, not proteins. Scales (as tabulated in ExPASy ProtScale): Kyte–Doolittle hydropathy (1982); Eisenberg normalised consensus hydrophobicity (1984); Hopp–Woods hydrophilicity (1981); Grantham polarity (1974); Janin interior-to-surface transfer energy (listed here as surface accessibility); Vihinen flexibility (normalised B-values, 1994); Zimmerman bulkiness (1968); and the molecular weight of the free amino acid in Da.

3D structure

With a UniProt accession the model URL is taken from the AlphaFold DB prediction API and the PDB file is downloaded; otherwise your uploaded PDB is used. The first chain is read (one residue per Cα atom) and aligned to the loaded sequence by exact substring match; if neither contains the other, residues are coloured 1:1 and a warning is shown. The right panel uses the property window size from the Sequence coverage tab. Rendering: 3Dmol.js.

Detection vs. properties

For each scale, the mean per-residue value (window 1) over covered residues and over uncovered residues, and their difference. The bars show where each mean lies between the scale's lowest and highest value in this protein. This is descriptive only: no statistical test is made, and detection also depends on enzyme, instrument and search settings.

PTM sites

Modifications written in ProForma are placed on the protein at the matched position; N- and C-terminal modifications sit on the first and last residue of the peptide. Modifications without a single residue (unknown position, range, localisation group, labile, global) are listed separately and are not drawn. Mass-delta modifications are matched against tacular's bundled UNIMOD table (UNIMOD_LOOKUP.query_mass) within half a unit of the last decimal written (at least 0.01 Da, at most 0.5 Da); residue specificity is not checked. Spec: ProForma.

Digest check

The protein is digested in silico with peptacular.digest_records() (chosen protease, up to the given number of missed cleavages, peptides within the length window). Best possible coverage is the share of residues inside at least one of those peptides; residues outside all of them are unreachable with these settings. A theoretical peptide counts as observed when its sequence equals the unmodified sequence of one of your matched peptides. Masses are neutral monoisotopic masses of the unmodified peptide (peptacular.mass).

Share links

The protein, peptide list and view settings are stored LZ-compressed in the part of the URL after #. Browsers do not send that part to the server, so long peptide lists fit and stay private. Use Copy share link.