Homology modeling structure prediction

The electron transfer, as observed in vitro between these two proteins, is believed to occur through the formation of a speciflc complex between the two proteins, being the interaction mainly electrostatic in nature. The nature and properties of the protein-protein complex (stoichiometry, interaction sites, association constants) were probed. Integration and correlation of the experimental results obtained from magnetic resonance studies on protein-protein titrations with the available structural and biochemical data is presented. A structural model for a hypothetical ternary complex, formed between one molecule of flavodoxin and two molecules of cytochrome C3, is proposed using the available X-ray structures of the isolated proteins and, when required, model structures predicted by homology modeling. 

TF Flavel. Predicting the structure of the fiavodoxm from Eschericia coli by homology modeling, distance geometry and molecular dynamics. Mol Simul 10 175-210, 1993. 

Building sequence profiles or Hidden Markov Models to perform more sensitive homology searches. A sequence profile contains information about the variability of every sequence position, improving structure prediction methods (secondary structure prediction). Sequence profile searches have become readily available through the introduction of PsiBLAST [4]... 

The aim of the second dimension depth is to consider protein 3D-stmctures to uncover structure-function relationships. Starting from the protein sequences, the steps in the depth dimension are structure prediction, homology modeling of protein structures, and the simulation of protein-protein interactions and ligand-complexes. 

Predicting a likely conformation or fold of a particular region of a protein with less or no sequence similarity to protein structures recorded in the PDB, is the main challenges for homology modeling of proteins. 

Cruciani et al. [92] have developed the program Metasite for the prediction of the site of oxidative metabolism by CYP450 enzymes. Metasite uses GRID molecular interaction fields to fingerprint both structures of CYP450s (from homology models or crystal structures) and test substrates and then matches the fields. Zhou et al. [93] showed that Metasite was able to correctly predict the site(s) of metabolism 78% of the time for 227 CYP3A4 substrates. Caron et al. [94] used Metasite to predict the oxidative metabolism of seven statins. 

For a reference supporting the importance of Arg358 and Phe357 for DPP-4 selectivity, see Rummy, C. and Metz, G. (2007) Homology models of dipeptidyl peptidases 8 and 9 with a focus on lop predictions near the active site. Proteins Structure, Function, and Bioinformatics, 66, 160-171. 

---