Secondary and Tertiary Structure Prediction of Proteins: A Bioinformatic Approach - Complex System Modelling and Control Through Intelligent Soft Computations - page 540

fast and inexpensive way to build a close approximation of a structure from a
sequence without involving the time and costs of experimental procedures. Fold
Recognition (FR) was reserved for methods which did not rely on sequence
searching and where the sequence identity between target and template was below
the so-called

30 %. The rationale behind the
threading method is that total number of experimentally solved 3D structure
deposited in PDB database doesn

“ twilight zone ” spanning between 25

-

t have a new fold. The nature has limited number
of basic folds which form the framework of most of the protein structures available
in PDB. Generally, similar sequence implies similar structure but the reverse is not
true. Similar structures are often found for proteins for which no sequence similarity
to any known structure can be detected (Floudas et al. 2006 ). Using fold recognition
or threading, we are able to identify proteins with known structures that share
common folds with the target sequences. Fold recognition methods work by com-
paring each target sequence against a library of potential fold templates using energy
potentials and/or other similarity scoring methods. For such comparison, we

'

rst

need to de ne a library of potential folds. Once the library is de ned, the target

sequence will be fitted into each library entry and an energy function is used to

evaluate the fit between the target sequence and the library entries to determine the

best possible templates. The template with the lowest energy score is then assumed
to best fit the fold of the target protein.

Fold recognition methods also includes various properties of structural envi-
ronment of the amino acid residue. Structural environments are more conserved
than the actual type of residue, therefore in the absence of homology, a fold could
be predicted by measuring the compatibility of a sequence with template folds in
terms of amino acid preferences for certain structural environments. These amino
acid preferences for structural environment provide suf

cient information to choose
among the folds. The amino acid preferences for three main types of structural
environment comprise of the solvent accessibility, the contact with polar atoms and
the secondary structure. The main limitation of this method is high computational
cost, since each entry in the whole library of thousands of possible folds needs to be
aligned in all possible ways to select the fold(s). Another major bottleneck is the
energy function used for the evaluation of alignment. It is not reasonable to expect
to find the correct folds in all cases with a single form of energy function. Few

popular web servers for modeling the protein structure by threading method are
listed in Table 4 .

3.3 Homology Modeling or Comparative Modeling

Comparative or homology protein structure modeling builds a three-dimensional
model for a protein of unknown structure (the target) based on one or more related
proteins of known structure. The necessary conditions for getting a useful model are