IDRBind is a protein interface predictor for binding sites of intrinsically disordered protein regions (IDRs), ranging from peptide motifs (i.e., short linear motifs) to molecular recognition features (MoRFs). Differentiating IDRBind from other interface predictors is its emphasis on binding sites of MoRFs, which are long interaction mediating elements within IDRs.

Protein-protein interactions have a fundamental role in most biological processes, and many computational methods have been created for the prediction of these interaction interfaces based on protein structure. Most of these methods have been trained and tested on classical globular-globular protein interactions. However, a large fraction of interactions are mediated by MoRFs, which play an especially prominent role in signaling and regulation. IDRBind is a method built to predict MoRF-binding sites on structures of folded proteins. It makes use of information from residue composition, residue evolutionary conservation, residue relative accessible surface area (rASA), surface geometry, estimated B-factor and electrostatics. The similarities in characteristics between binding sites of MoRFs and shorter peptides (e.g., SLiMs) make IDRBind a suitable prediction method for both.

Please see FAQ for more usage information.

A pre-computed dataset of human proteins is available, but more careful selection of input structure would provide better results. See included reademe.txt. https://drive.google.com/file/d/1amvuuLUGf77oqZ_BdN_vGagbwni_tRCE/view?usp=sharing

Reference for IDRBind:

E.T.C. Wong, J. Gsponer “Predicting protein-protein interfaces that bind intrinsically disordered protein regions” J. Mol. Biol. (2019). (Accepted)

PDB file - Upload the protein(s) coordinates in standard PDB file format. It is recommended to have properly terminated chains. When multiple models of the protein are available, only the first model in the file will be used.

PDB Chain ID(s) - You must specify one or more protein chains matching those found in the PDB file (e.g., AB).

FASTA file - We also require a sequence file in the FASTA format where each sequence is labeled with chain ID matching those in the PDB file (not MoRF sequence).

Job Name - Optional label for your job to identify it in the public jobs queue and its individual job page.

Email Address - Optional email address your you to be notified at when your job has entered the process queue, begun processing, and finished processing.

Keep this job private - Optionally prevent this job from showing up in the public jobs queue.

NOTE: If job is set to private, remember to save the URL of the job's page or to supply an email so as to not to accidentally lose access to your job!