Capturing Protein Sequence Data: UniProt Knowledgebase (UniProtKB) submissions and Journal Scanning

Most sequences in UniprotKB are derived from the translation of nucleotide sequence, but we also add sequences obtained through direct sequencing of peptides by Edman degradation and tandem mass spectrometry. This allows us to capture protein sequences for which no nucleotide sequence is available, and can also confirm that a predicted protein exists and elucidate the post-translational processing it undergoes. Scientists submit protein sequences to UniProtKB via the web-based submission tool "SPIN":http://www.ebi.ac.uk/swissprot/Submissions/spin. Curators use a range of sequence analysis tools to check the submission and frequently discuss the sequence and supporting evidence with the submitter. Submitted entries can be kept confidential for a limited period at the submitter's request. UniProtKB submission is complemented by the journal scan project that monitors the scientific literature for papers containing directly sequenced proteins. UniProtKB entries are then created for those sequences that are not yet in the database and have not been submitted via SPIN.