Chemistry Reference
In-Depth Information
receptor) Monte Carlo-based docking which is actually an extension to homology
modeling.
It is necessary to take into consideration comparisons with corresponding
performance on crystal structures, enrichment factors, docking poses, availability
of template structures, the docking program, the ligand library, choice of decoy
library, ligand preparation, target-dependency as well as the rigid receptor
approximation.
For accurate ligand-protein docking on homology models, it is necessary to model
accurately the binding sites and account appropriately for receptor flexibility.
In
silico
homology modeling thus appears to be a promising and cost-effective
approach established in structure-based virtual screening protocols. There is still
need to develop better methods in order to accurately model target-binding sites
and improve refinement at the backbone and side chain levels. The adequate
refinement process may offset a low target-template sequence similarity [148].
Some useful tools that have been used for homology modeling are: 1) PDB
(www.pdb.org), (Database); 2) SWISS-PROT (www.ebi.ac.uk/swiss.prot),
(Database); 3) BLAST (www.ncbi. nlm.nih.gov/BLAST/), (for identifying structural
templates/sequence alignment); 4) Modeller (www.guitar.rockfeller.edu/modeller/
modeller.html), (for model building); 5)WHATCHECK (www.cmbi.kun.nl.swift/
whatcheck/), (for model validation).
Some authors have used docking with knowledge-based scoring functions and
homology models to obtain good success rates of pose prediction [141]. Other
workers have used template-target alignment methods and evolutionary distances
with a set of homology models and crystals structures to produce robust docking
results [143]. Another approach involves homology models with binding site
optimization (induced-fit docking) which yields improved docking results [149].
Homology models can also be used for binding into a number of important
receptors [91, 142-149].
FRAGMENT-BASED
FBDD (Fragment-based drug design) yields a very promising approach for
optimization and discovery of lead compounds. The experimental approach is
