Tutorial: How to Submit A Set of Protein Interactions

* Click on Submit a new job

* Enter the name of the set you are about to submit.
Example: Network1, Rhodopsin_signalling.

* Upload your sequence file, in fasta format . see example
IMPORTANT: for an optimal use of the resources provided by SAPIN, it is required to use uniprot accession numbers. Using alternative protein/gene id may result in an incomplete analysis of the domain-motif interactions.

To facilitate the conversion of your favorite protein IDs into uniprot accession numbers, you can visit the ID mapping tool provided on the UNIPROT website.

* Upload your interaction file, which is a two column file, where each line contains an interaction between two proteins (column 1 and column 2). see example
IMPORTANT: for an optimal use of the resources provided by SAPIN, it is required to use uniprot accession numbers.
Additionally, the proteins provided in the interaction file should have their corresponding sequence in the fasta file identified by the exact same ID.

* Click on "Submit" to start the structural analysis of your interaction set.

* Refresh the page to check the status of your job. Alternatively, you can create an account to receive a mail when the job finishes. The job has finished correctly when both start and end dates are given, as shown below.

Accessing the Results

* To access the results, you can either click on the dataset name in the job list or by downloding the zip file containing all the results in text format files.

Looking at the Sequence Analysis Results

The protein sequences provided by the user are in a first step analysed using various programs and databases:

     *PFAM  - Domain composition

     *Psipred - Prediction of secondary structures
        line 1 : C = Coil ; H = Helix ; E = Sheet
        line 2 : Confidence score
        The color code highlights the secondary structures elements (alpha helices and beta sheets)

     *Disopred  - Prediction of disordered regions
        line 1 : . = residue predicted as non-disordered ; * = residue predicted as disordered
        line 2 : Confidence score

     *PhosphoELM  - Database of Phosphorylation sites

     *NetPhorest - Classifiers of non-linear motifs involved in phosphorylation-depending signalling

     *Motif-binding domains based on regular expression searches derived from ELM and from Stein and Aloy (2008) .

     *Scansite - Motifs phosphorylated by kinases or binding SH2, 14-3-3 or PDZ domains.
        We note that Partners represent potential binding domains.

Looking at the Network

The analysis of competition for binding is displayed using the Cytoscape Web application, which allows the user to interact with the network. This network results from the structural analysis based on domain-domain interactions. To display this information, we show the proteins (RECTANGLES), their domains involved in an interaction (CIRCLES). If a domain is involved in the binding with two domains, the interactions are evaluated by structural superimposition and backbone clash calculation. If a domain can interact simultaneously with other domains, we indicate as a green node color ('AND'); if any interactions are mutually exclusive, the node is colored as red ('XOR').
If the network window is empty, it means that they were not enough structural information to carry out the competition analysis.

help_dataset_name.png9.21 KB
help_fasta_file.png16.99 KB
help_interaction_file.png16.76 KB
help_job_description_1.png50.52 KB
help_job_description_2.png37.82 KB