Help
1. Summary
 |
|
|
Pathopred is a pathogenicity predictor for amino acid substitutions, utilising a deep convolutional neural network trained on a large amount of variants in human proteins, incorporating evolutionary information, sequence information, as well as structural information.
|
2. Usage
|
|
|
Sequence input to the server is an amino acid sequence in FASTA format. The users can submit
queries by either pasting one or several sequences in the text-area
provided, or, alternatively, uploading a file for batch submission of
sequences. BLAST will be used to find sequence homologues for each sequence.
Example input:
>P26439
MGWSCLVTGAGGLLGQRIVRLLVEEKELKEIRALDKAFRPELREEFSKLQNRTKLTVLEG
DILDEPFLKRACQDVSVVIHTACIIDVFGVTHRESIMNVNVKGTQLLLEACVQASVPVFI
YTSSIEVAGPNSYKEIIQNGHEEEPLENTWPTPYPYSKKLAEKAVLAANGWNLKNGDTLY
TCALRPTYIYGEGGPFLSASINEALNNNGILSSVGKFSTVNPVYVGNVAWAHILALRALR
DPKKAPSVRGQFYYISDDTPHQSYDNLNYILSKEFGLRLDSRWSLPLTLMYWIGFLLEVV
SFLLSPIYSYQPPFNRHTVTLSNSVFTFSYKKAQRDLAYKPLYSWEEAKQKTVEWVGSLV
DRHKETLKSKTQ
Variant input to the server is written on the format of reference AA, position,
altered AA, e.g. A509G. The variants should be specified below their corresponding sequence identifier starting with >.
Note that the sequence identifiers should be Uniprot IDs. If you do not have them on this format, you can use the Uniprot Retrieve/ID mapping tool to prepare them.
Example input:
>P26439
P186L
P222H
P222Q
|
3. Output
|
|
|
When the server has finished running a prediction, an output file with predictions will be generated. This file contains a prediction of whether a variant is pathogenic or not, as well as a prediction of whether the variant is severe or not in case of a pathogenic variant. A simple chart of the sequence entropy with the position of the variants marked will also be generated.
All files can be downloaded as a zip archive file for further analysis.
|
|
Furthermore, examples of results can be found here
|
4. References
5. Contact
|
|
|
Arne Elofsson group
Department for Biochemistry and Biophysics
The Arrhenius Laboratories for Natural Sciences
Stockholm University
SE-106 91 Stockholm, Sweden
Science for Life Laboratory
Box 1031, 17121 Solna, Sweden
|
|
|
