Homology/Comparative Modeling


1. Question to be solved:

Build a 3D structure of the following protein. This is known to be an RNA-binding protein. But the RNA-binding sites(residues) are not known.
>EWHA_PRACTICE
MAEVEEKGLLSYYFPGERIPDSIISQDGSIRLGPGLVFQKKDREEEIVVSKAGRLHQTGK
NAMLIDSRTKRYIPATNDQVIGQIISRFAEGYRVDIGSAHIAQLNALAFENVTRKSRPNL
NVGSLVYARVSLADRDMEPELECFDATTGKAAGYGELKNGYMITGLSLSHCRKLILPKNT
LLQTLGSYIPFEIAVGMNGRVWVNSENLSTTVLICTAIRNCEFMSDEEQIKYCKDLIKKL

Procedure for Homology/Comparative Modelling

  1. Save the amino acid sequence above to your PC.
  2. Find the template for homology/comparative modelling

    Use the template linked here in this practice. This is the coordinate of a protein with PDB ID 2PEH (chain B). Save the linked file to your PC.
    In the real world, you have to use BLAST/FASTA to search for the best template.

  3. Build the alignment for homology/comparative modeling

    Get amino acid sequence of the template (This is sometimes very tricky). Then perform pairwise alignment of the two sequences. The following website can deal with all the trick. http://cib.cf.ocha.ac.jp/EWHA/buildalign.html
    Save the alignment to your PC.
    In the real world, you have to use ClustalW, ALAdeGAP or other related tools.

  4. Build a coordinate of the target sequence.

    Let us use,

  5. Evaluate the accuracy of the modelled structure

2. Use different website to build an alignment and build a different 3D structure, based on the aligment.


update: 9th, Oct., 2013
This website is used for practicing homology/comparative modelling for graduate students.
Kei Yura, Center for Informational Biology, Ochanomizu University, Japan