Prediction of viral protease inhibitors using proteochemometrics approach.

Being widely accepted tools in computational drug search, the (Q)SAR methods have limitations related to data incompleteness. The proteochemometrics (PCM) approach expands the applicability area by using description for both protein and ligand structures. The PCM algorithms are urgently required for the development of new antiviral agents. We suggest the PCM method using the TLMNA descriptors, combining the MNA descriptors of ligands and protein sequence N-grams. Our method was validated on the viral chymotrypsin-like proteases and their ligands. We have developed an original protocol allowing us to collect a comprehensive set of 15 protein sequences and more than 9000 ligands from the ChEMBL database. The N-grams were derived from the 3D-based alignment, accurately superposing ligand-binding regions. In testing the ligand set in SAR mode with MNA descriptors, an accuracy above 0.95 was determined that shows the perspective of the antiviral drug search in virtual chemical libraries. The effective PCM models were built with the TLMNA descriptor. The strong validation procedure with pair exclusion simulated the prediction of interactions between the new ligands and new targets, resulting in accuracy estimation up to 0.89. The PCM approach shows slightly lower accuracy caused by more uncertainty compared with SAR, but it overcomes the problem of data incompleteness.

The method predicting interaction between protein targets and small-molecular ligands with the wide applicability domain.

Prediction of protein-ligand interaction is necessary for drug design, gene regulatory networks investigation, and chemical probes detection. The existing methods commonly demonstrate high prediction accuracy for the particular groups of protein and their ligands. We developed an approach suited for the wider applicability and tested it on three dataset types significantly differing by protein homology. The study included three typical scenarios of assessing the target-ligand interaction: 1st - predicting protein targets by ligand structures' comparisons; 2nd - predicting ligands by target sequences' comparisons; 3rd - predicting both the uncharacterized targets and ligands with the fuzzy coefficients based on ligand comparisons. The 1st scenario implemented showed a high prediction accuracy of 0.96-0.99, providing fuzzy coefficients of target-ligand interactions in the 3rd scenario. Testing by 2nd scenario displayed the accuracy of 0.97-0.99 for predicting within the particular protein families, sets non-ordered by protein homology, and accuracy higher than 0.90 for most HIV sets, each presenting the close mutant proteins differing by point substitutions. The 3rd scenario displayed that fuzzy classification can reveal reasonable accuracy 0.86-0.94 at simulated data incompleteness. Thus, our approach provides high prediction accuracy with the wide applicability domain, including data differing in heterogeneity and completeness.

Spatial features of proteins related to their phosphorylation and associated structural changes.

Protein phosphorylation is widely used in biological regulatory processes. The study of spatial features related to phosphorylation sites is necessary to increase the efficacy of recognition of phosphorylation patterns in protein sequences. Using the data on phosphosites found in amino acid sequences, we mapped these sites onto 3D structures and studied the structural variability of the same sites in different PDB entries related to the same proteins. Solvent accessibility was calculated for the residues known to be phosphorylated. A significant change in accessibility was shown for many sites, but several ones were determined as buried in all the structures considered. Most phosphosites were found in coil regions. However, a significant portion was located in the structurally stable ordered regions. Comparison of structures with the same sites in modified and unmodified states showed that the region surrounding a site could be significantly shifted due to phosphorylation. Comparison between non-modified structures (as well as between the modified ones) suggested that phosphorylation stabilizes one of the possible conformations. The local structure around the site could be changed due to phosphorylation, but often the initial conformation of the site surrounding is not altered within bounds of a rather large substructure. In this case, we can observe an extensive displacement within a protein domain. Phosphorylation without structural alteration seems to provide the interface for domain-domain or protein-protein interactions. Accounting for structural features is important for revealing more specific patterns of phosphorylation. It is also necessary for explaining structural changes as a basis for regulatory processes.

Prediction of amino acid positions specific for functional groups in a protein family based on local sequence similarity.

The exchange of single amino acid residue in protein can substantially affect the specificity of molecular recognition. Many protein families can be divided into the groups based on specificity to recognized ligands. Prediction of group-discriminating residues within the certain family is extremely necessary for theoretical studies, enzyme engineering, drug design, and so on. The most existing methods use the multiple sequence alignment. They have the limitations in prediction accuracy due to the family sequence divergence and ligand-based grouping. We developed a new method SPrOS (Specificity Projection On Sequence) for estimating the specificity of residues to user-defined groups. SPrOS compares the sequence segments from the test protein and training proteins. Contrary to other segment-comparison approaches extracting the string motifs, SPrOS calculates the scores for single positions by the similarity of their surroundings. The method was evaluated on the simulated sequences and real protein families. The high-prediction accuracy was achieved for simulated sequences, in which SPrOS detected specific positions not predicted with the alignment-based method. For bacterial transcription factors (LacI/GalR) clearly divided into functional groups, the predicted specific residues corresponded to the published experimental data. In a more complicated case of protein kinases classified by inhibitor specificity, the positions predicted with high significance were located in ligand-binding areas. As the ligand specificity is not necessary coincided with phylogeny, evolutionary-coupled mutations could disturb the detection of ligand-specific residues. Excluding proximate homologs of the test protein kinase from the training set, we improved the prediction of the ligand-specific residues. The SPrOS is available at http://www.way2drug.com/spros/

Prediction of amino acid residues participated in substrate recognition by cytochrome P450 subfamilies with broad substrate specificity.

Cytochromes P450 comprise a large superfamily and several of their isoforms play a crucial role in metabolism of xenobiotics, including drugs. Although these enzymes demonstrate broad and cross-substrate specificity, different cytochrome P450 subfamilies exhibit certain selectivity for some types of substrates. Analysis of amino acid residues of the active sites of six cytochrome subfamilies (CYP1А, CYP2А, CYP2С, CYP2D, CYP2E and CYP3А) enables to define subfamily-specific patterns that consist of four residues. These residues are located on the periphery of the active sites of these cytochromes. We suggest that they can form a primary binding site at the entrance to the active site, defining cytochrome substrate recognition.

Rational design of macrolides by virtual screening of combinatorial libraries generated through in silico manipulation of polyketide synthases.

Bacterial secondary metabolites display diverse biological activities, thus having potential as pharmacological agents. Although most of these compounds are discovered by random screening, it is possible to predict and re-design their structures based on the information on their biosynthetic pathways. Biosynthesis of macrolides, governed by modular polyketide synthases (PKS), obeys certain rules, which can be simulated in silico. PKS mode of action theoretically allows for a huge number of macrolides to be produced upon combinatorial manipulation. Since engineering of all possible PKS variants is practically unfeasible, we created Biogenerator software, which simulates manipulation of PKS and generates virtual libraries of macrolides. These libraries can be screened by computer-aided prediction of biological activities, as exemplified by analysis of erythromycin and macrolactin libraries. This approach allows rational selection of macrolides with desired biological activities and provides instructions regarding the composition of the PKS gene clusters necessary for microbial production of such molecules.