Submit - Step 1

19/10/2008: The Meta Server has been moved to a new hardware. Please report any 'suspicious' (in)activity.

The Meta Server is kept running by the concerted effort of its users and maintainers. Please report bugs and unusual behaviour to the maintainers.
We receive your message even if we can not respond immediately - it often gets quite busy here. Thank you!

Modeller is available to all our users who possess a valid Modeller licence key. Click on the [model] link at the right end of an alignment on the 3D-Jury page to get to the Modeller submission page. Your model will be e-mailed to you.

Usage hints

• Submit domains separately
• Remove coiled coil regions
• Jobs queued for more than 7 days for servers with queue>30 are skipped
• Use is limited to 30 jobs per week per domain
• Contact us in case of problems with interpretation of results
• Some servers return only models, no alignments (target sequence is shown)
• Results published on this server are public and can not be used for patenting

The Structure Prediction Meta Server provides access to various fold recognition, function prediction and local structure prediction methods. The Server takes the amino acid sequence of the query protein, the reference name for the prediction job, and the E-mail address as input. The E-mail address is used only for notification about errors during the execution of the job. The query sequence and the reference name are placed in the process queue. The Meta Server accepts only sequences, which have not been submitted before. In case of duplicate sequences the second user will be notified with a link to the previous submission. Sequences longer than 800 amino acids are not accepted by some services. The internal SQL database offers the possibility to find any previous jobs processed by the Meta Server using "regular expressions" addressing field like E-mail, Job Name and the host name, from which the job was initiated. Each server has its own process queuing system managed by the Meta Server. All results of fold recognition servers are translated into uniform formats. The information extracted from the raw output of the servers includes the PDB codes of the hits, the alignments and the similarity (reliability) scores specific for every server. Mapping of the hits to the SCOP and FSSP classifications are made either using known PDB representatives or alignment of the template sequence with the databases of proteins in both classifications. The secondary structure assignments for all hits are taken from the mapped FSSP (red for helices and blue for strands). Underscored amino acids indicate the first residue after an insertion in the template sequence. The Meta server provides translation of the alignments in standard formats like FASTA, PDB or CASP. The Meta Server is coupled to consensus servers. They provide jury predictions based on the results collected from other services. Not all fold recognition servers are used by the jury system. The data stored on the meta server is available through http://meta.bioinfo.pl/data/JOBID/. Jobs older than 2 months are not shown. The Meta Server is only a set of programs aimed to process and manage biological data, while the predictive power of the service comes from (mostly) remote prediction providers.

Please cite the prediction servers and 3D-Jury: Ginalski K, Elofsson A, Fischer D, Rychlewski L. "3D-Jury: a simple approach to improve protein structure predictions." Bioinformatics. 2003 May 22;19(8):1015-8. [Library]