This is a gateway to various methods for protein structure prediction.
If you submit a single protein sequence or a multiple sequence alignment
(hereafter called a "target"), you should obtain the following predictions:
Primary structure: domains identified by HmmPfam and HHSearchCDD
Secondary structure: helical/extended/other conformation
(H/E/-) sspro4, cdm, psipred, fdm, jnet, prof, gor, spine, sable, pssfinder, sspred, sspal, nnssp, ssp, NetSurfP, SOPRANO
Transmembrane helices: Phobius, MEMSAT3, TMpred, HMMTOP, DAS, MINNOU, OCTOPUS, TMHMM2.0
Disordered regions: DisEMBL,RONN, IUPRED, POODLE-L, GLOBPLOT, DISPROT(VSL2), PrDOS, DISOPRED2, Spritz, disPRO, Pdisorder, Spritz, iPDA, POODLE-S (calculated consensus is CASP8 winning method in the category).
Disulfide bonds: DiANNA, DISULFIND, DBCP and DIpro2.0
Nucleic acid binding residues in proteins: BindN, BindN+, DISIS, DP-Bind, PPRInt, and our meta-predictor for RNA-binding residues (cons3best) - for details see here
Tertiary structure: alignments of the target sequence to sequences of proteins
with known structures, identified by protein fold-recognition (FR) methods.
The Pcons consensus server will evaluate to which extent the FR alignments agree
with each other and if a particular fold can be singled out.
From FR alignments you can automatically generate crude three-dimensional models
(without variable loops). You can also select a subset of alignments that
belong to a particular fold and submit them to our FRankenstein server, which
will splice and recombine the crude models to produce a cute little monster
protein model that hopefully resembles the true structure.
Models produced by our server are scored by COLORADO3D. Use "color by
temperature factor" option in RASMOL to visualize regions likely to be correct
(blue) and those that may be wrong (red). Please remember that all theoretical
models must be taken with a grain of salt - and even the most confident ones
should be carefully verified.
Whenever disorder prediction used: Kozlowski LP, Bujnicki JM. MetaDisorder: a meta-server for the prediction of intrinsic disorder in proteins. BMC Bioinformatics. 2012 May 24;13(1):111.
For meta-predictor of RNA-binding residues (cons3best): Puton T, Kozlowski L, Tuszynska I, Rother K, Bujnicki JM. Computational methods for prediction of protein-RNA interactions. J Struct Biol. 2012; 179(3):261-8
Main citation: Kurowski MA, Bujnicki JM. GeneSilico protein structure prediction meta-server. Nucleic Acids Res 2003; 31: 3305-3307
Important technical note:
This site is fully functional with W3C DOM compliant browsers only. Users of
older browsers, for example Netscape 4.x, may experience problems. There may be
some glitches with MS Internet Explorer, especially versions < 6.0.
We recommend using browsers based on the recent release of the "Gecko" engine
(for example Firefox) or Safari.