iMOT is an automated method of identifying conserved spatially interacting regions across proteins. Signatures of proteins are derived based on spatial interactions among sequentially conserved regions. Interaction of the conserved stretches are evaluated based on pseudo-energies considering sterics, electrostatics and hydrophobics.

Flow chart of the methodology adapted for identifying conserved regions or interacting motifs by various methods.

Spatially interacting motifs of individual query sequences are evaluated for a given alignment. Each of the motifs is mapped onto the alignment. Motifs sharing equivalent positions on the structure or similar regions in the alignment represent the sets of common interacting motifs. These common interacting motifs are tabulated and shown on the provided alignment.

The common interacting motifs which reside at equivalent regions of the provided query structures have been mapped onto the alignment. Only those motifs have been mapped which are present in atleast two sequences.

The amino acid substitutions at each alignment position were scored by referring to a symmetric amino acid exchange matrix (Johnson and Overington 1993) derived from several alignments of homologous protein structures. Regions that contain at least two residues (with a score more than 50) in a window of five were considered to be conserved.

Interactions between conserved stretches have been evaluated based on pseudo-energies. Pseudo-energies include components of electrostatic, sterics and hydrophobic interactions. Electrostatic interaction is calculated based on the Debye-Huckel expression similar to the approach adapted by Crichton and co-workers [Dimitrov & Crichton 1997]. The hydrophobic [Novotny et al. 1997] and steric interactions [Lomize et al 2002] are evaluated based on previously published approaches by other groups.

Start : Start position of the motif in a given alignment

End : End position of the motif in a given alignment

Residue number in pdb file::

Start : Number assigned (in PDB file) to the first residue of the motif

End : Number assigned (in PDB file) to the last residue of the motif

Input sequence should be in PIR format. User can supply structural alignments of family or superfamily to find motifs for that family or superfamily. User has to give only one sequence to find iMOT for its structural homologue or to find sequentially conserved motifs. One can get the structural alignment from PASS2, CAMPASS and HOMSTRAD.

To identify homologues for a given sequence, we employ BLAST for searching sequence databases. The search can be performed on the Protein Databank (PDB), the SWISS-PROT and the Non redundant sequence databases (NR). Though a search against a large database may provide a better opportunity to select homologues, the searches are time consuming. We, therefore, warn that searches against NR database may be long!

Non redundant: All Non redundant GenBank CDS translations+RefSeq Proteins+PDB+SwissProt+PIR+PRF

SWISS-PROT : Last major release of the SWISS-PROT protein sequence database (no updates)

PDB sequence database : Sequences derived from the 3-dimensional structure from Brookhaven Protein Data Bank

The E-value or the expectation value is a statistical significance threshold for reporting matches against database of sequences. If the statistical significance ascribed to a match is greater than the EXPECT threshold, the match will not be reported. Lower EXPECT thresholds are more stringent, leading to fewer chance matches being reported. Increasing the threshold shows less stringent matches.

* Paste structural alignments in PIR format. Please do not submit alignment containing more than 5 proteins at a time. Very large inputs may lead to refusal of the service. Since blasts are time consuming, you may have to wait for few minutes.

* Select database from the menu. SwissProt is recommended. NR will be time consuming.

* Select e-value. Default value is found good for a number of cases, and is recommended.

* Select the prediction method.

* Press the "Submit query" button.



Output format:

Find iMOT for a sequence:

* Paste your sequence in PIR format.

* Select database from the menu. SwissProt is recommended. NR will be time consuming.

* Select e-value. Default value is found good for a number of cases, and is recommended.

* Select the prediction method.

* Press the "Submit query" button.

* If your sequence retrieves any structural homologue, iMOT server finds the interacting motifs for the structural homologue and aligns the sequence with it.

Output format: Find sequentially conserved region(s) for a sequence:

* Paste your sequence in PIR format.

* Select the prediction method.

* Press the "Submit query" button.

Output format:

 E > -50        ---- Weak interaction

 -125 > E <= -50 ---- optimal interaction 

 E <= -125       ---- strong interaction