Homology Modeling: Generating Structural Models to Understand Protein Function and Mechanism - Computational Modeling of Biological Systems - page 98

NMR spectroscopy is not tractable. Comparative modeling, or homology modeling,
exploits the fact that two proteins whose sequences are evolutionarily connected
display similar structural features [ 3 ]. Thus, the known structure of a protein
(template) can be used to generate a molecular model of the protein (query) whose
experimental structure is not known.
The applicability of comparative modeling in structural biology has been vali-
dated by the observations of several groups, e.g., that a limited number of protein
folds are observed in nature [ 4 , 5 ] and that nature is able to reuse similar folds for
diverse protein functions [ 6 ]. Thus, several researchers have used the already avail-
able breadth of structural information to build structural models of many proteins
whose experimental structures have not been determined. For example, ModBase
[ 7 ] and SWISS-MODEL [ 8 ], repositories of comparative models generated using
automated protocols, have structural models for 3.4 million and 2.2 million unique
sequences respectively; for comparison, the repository for experimental structures,
protein data bank (PDB [ 9 ]) has 67,728 experimental structures. The burgeoning
number of structural models in repositories such as ModBase and SWISS-MODEL
reflects the usefulness of comparative modeling in significantly closing the gap
between the number of known sequences and known structures. To further close
this gap, the protein structure initiative [ 10 ] aims to determine the experimental
structures for representative members of protein families that do not yet have any
structural templates in the PDB.
Structural models generated by homology modeling can be of direct medical
and biological relevance. Structural models can be used to predict the effects of
single nucleotide polymorphisms uncovered from genome-wide association studies,
helping to delineate the molecular etiology of genetically transmitted diseases [ 2 ].
Homology-based structural models have already been used widely in in silico
drug screening [ 11 - 13 ]. For biological experiments, structural models can be
used to design mutations that lead to specific changes in the function or stability
of the modeled protein [ 14 , 15 ]. Importantly, homology models can be used as
starting models for molecular replacement in X-ray crystallography [ 16 ], leading
to better experimental structures. Furthermore, these structural models can be used
in conjunction with methods such as FRET that provide interresidue distances and
for mapping residue-level experimental data, such as accessibility measured through
EPR [ 17 ] and H-D exchange mass spectrometry [ 18 , 19 ].
Thus, to better understand the function and mechanism of a given protein of
unknown structure, researchers can generate structural models using comparative
modeling. In this chapter, we discuss the process of generating a homology-based
structural model of a protein of interest. In particular, we focus on the critical con-
trols and tests to be used at each step of model building to ensure that the final model
is physically and biologically reasonable and, most importantly, to determine the
extent to which the given model can be used in interpretations of experimental data.
Comparative modeling involves several steps, such as identification of the template,
sequence threading, processing insertions and deletions, model optimization, quality
control, and finally, model interpretation (Fig. 1 ,Table 1 ). We discuss each of these
steps in the following sections.