Submitting a job

The user can submit one or more single residue variation/s occurring on one or more protein sequence/s. Each unique sequence has to be provided in a fasta-like format. The header of each sequence must contain a protein identifier followed, after a space, by the list of variations separated by commas, in the format [wildtype residue][position][variant residue] (see example below).

        >O43776 R11P,T17M,K60E,G132C,R322L,L350P,D356A,A422T,T459I,G509S,R545C
        MVLAELYVSDREGSDATGDGTKEKPFKTGLKALMTVGKEPFPTIYVDSQKENERWNVISK
        SQLKNIKKMWHREQMKSESREKKEAEDSLRREKNLEEAKKITIKNDPSLPEPKCVKIGAL
        EGYRGQRVKVFGWVHRLRRQGKNLMFLVLRDGTGYLQCVLADELCQCYNGVLLSTESSVA
        VYGMLNLTPKGKQAPGGHELSCDFWELIGLAPAGGADNLINEESDVDVQLNNRHMMIRGE
        NMSKILKARSMVTRCFRDHFFDRGYYEVTPPTLVQTQVEGGATLFKLDYFGEEAFLTQSS
        QLYLETCLPALGDVFCIAQSYRAEQSRTRRHLAEYTHVEAECPFLTFDDLLNRLEDLVCD
        VVDRILKSPAGSIVHELNPNFQPPKRPFKRMNYSDAIVWLKEHDVKKEDGTFYEFGEDIP
        EAPERLMTDTINEPILLCRFPVEIKSFYMQRCPEDSRLTESVDVLMPNVGEIVGGSMRIF
        DSEEILAGYKREGIDPTPYYWYTDQRKYGTCPHGGYGLGLERFLTWILNRYHIRDVCLYP
        RFVQRCTP
    

Two example inputs can be directly pasted in the input window by clicking on the blue buttons. In alternative to paste variations into the input window, the user can upload a file, formatted as described above.

Before the job is submitted, a validation procedure is applied to the input:

• The protein ID cannot contain spaces or commas
• The protein sequence must contain only capital letters
• All wildtype residues specified in the variations must match the corresponding positions in the sequence
• All sequences must be unique. A job with two identical sequences is rejected
• The server accepts at most 1000 unique sequences in a single job

After submission, the user is redirected to a new page indicating that the job has been accepted. The page automatically refreshes every 10 seconds and it will display the status of your job (either queued or running). To access results at a later time, the URL of the page can be bookmarked.

Reading the results

Results are displayed in a tabular format, with a section for each submitted unique sequence. In each section, one line per variation is shown, with the following columns:

• Variant: Identifier of the variation
• Pathogenicity class: Predicted class, either Pathogenic or Benign
• Pathogenicity probability (Pathogenic: ≥ 0.5): Calibrated probability of a variation to be Pathogenic
• Reliability Index: an integer number in the range [0-10] where 0 and 10 correspond to the minimum and maximum confidence for the prediction. The Reliability Index (RI) is computed from the pathogenicity probability (P) with the formula RI = 20 * |P - 0.5|.