FASTA viewer

Reading the file. fastatacular.read_fasta() parses the file. Gzip, bzip2 and xz compression are detected from the file's first bytes, so the file name does not matter. UniProt-style headers (sp|P04637|P53_HUMAN … OS= OX= GN= PE= SV=) are split into database prefix, accession, entry name, protein name, organism, taxon ID, gene, protein existence level (PE) and sequence version (SV). Other KEY=value pairs are listed as extra fields. NCBI gi|…|ref|…| identifiers and plain identifiers are kept as they are.

Mass and formula. peptacular.mass(sequence, monoisotopic=True) and monoisotopic=False give the neutral monoisotopic and average mass of the unmodified sequence, including the terminal water (H₂O). The formula comes from peptacular.comp(), written C, H, N, O, S, P and then any other elements alphabetically. Element masses come from the tacular reference data. Sequences with ambiguous or non-amino-acid letters (for example X, B, *, or a nucleotide sequence read as protein) may have no mass; the reason is shown instead.

Composition. Counts of each letter in the sequence. The residue classes are this page's own grouping for colouring: hydrophobic A V I L M F W Y C; polar S T N Q; acidic D E; basic K R H; special G P. Other letters are counted but not classed.

Search. Runs in the page on the parsed entries. "All fields" matches the text in the identifier, accession, entry name, gene, protein name, organism and sequence (case-insensitive). "Peptide substring" counts every occurrence, including overlapping ones. "Regular expression" uses JavaScript regular-expression syntax on the upper-case sequence, e.g. N[^P][ST] for the N-glycosylation sequon.

Decoys. Entries whose identifier starts with DECOY_, decoy_, REV_, rev_ or XXX_ are flagged as decoys (fastatacular.is_decoy()). "Download target + decoy FASTA" calls fastatacular.write_decoy_fasta(): every target unchanged, then one decoy per target with the prefix added. It refuses a file that already holds decoys.

• Reverse: the sequence reversed.
• Pseudo-reverse: each stretch between K/R reversed with K and R kept in place, so tryptic decoy peptides keep the target peptides' lengths, compositions and masses.
• Shuffle: residues randomly permuted (redrawn if the shuffle gives back the target).
• de Bruijn: every distinct 3-mer of the target database gets one random replacement residue, so sequence repeated in the targets is repeated in the decoys (Moosa et al. 2020).
• Markov chain: a new sequence sampled from an order-2 Markov model trained on the human proteome.

References: Elias JE, Gygi SP. Target-decoy search strategy for increased confidence in large-scale protein identifications by mass spectrometry. Nat Methods 4:207–214 (2007). Moosa JM, Guan S, Moran MF, Ma B. Repeat-preserving decoy database for false discovery rate estimation in peptide identification. J Proteome Res 19:1029–1036 (2020). doi:10.1021/acs.jproteome.9b00555

CSV export. One row per listed entry with the columns of fastatacular's SequenceEntry.to_record().