The long coming of computational structural biology

Complex networks govern coiled-coil oligomerization--predicting and profiling by means of a machine learning approach

Understanding the relationship between protein sequence and structure is one of the great challenges in biology. In the case of the ubiquitous coiled-coil motif, structure and occurrence have been described in extensive detail, but there is a lack of insight into the rules that govern oligomerization, i.e. how many α-helices form a given coiled coil. To shed new light on the formation of two- and three-stranded coiled coils, we developed a machine learning approach to identify rules in the form of weighted amino acid patterns. These rules form the basis of our classification tool, PrOCoil, which also visualizes the contribution of each individual amino acid to the overall oligomeric tendency of a given coiled-coil sequence. We discovered that sequence positions previously thought irrelevant to direct coiled-coil interaction have an undeniable impact on stoichiometry. Our rules also demystify the oligomerization behavior of the yeast transcription factor GCN4, which can now be described as a hybrid--part dimer and part trimer--with both theoretical and experimental justification.

Structure-based identification of MHC binding peptides: Benchmarking of prediction accuracy

Identification of MHC binding peptides is essential for understanding the molecular mechanism of immune response. However, most of the prediction methods use motifs/profiles derived from experimental peptide binding data for specific MHC alleles, thus limiting their applicability only to those alleles for which such data is available. In this work we have developed a structure-based method which does not require experimental peptide binding data for training. Our method models MHC-peptide complexes using crystal structures of 170 MHC-peptide complexes and evaluates the binding energies using two well known residue based statistical pair potentials, namely Betancourt-Thirumalai (BT) and Miyazawa-Jernigan (MJ) matrices. Extensive benchmarking of prediction accuracy on a data set of 1654 epitopes from class I and class II alleles available in the SYFPEITHI database indicate that BT pair-potential can predict more than 60% of the known binders in case of 14 MHC alleles with AUC values for ROC curves ranging from 0.6 to 0.9. Similar benchmarking on 29,522 class I and class II MHC binding peptides with known IC(50) values in the IEDB database showed AUC values higher than 0.6 for 10 class I alleles and 9 class II alleles in predictions involving classification of a peptide to be binder or non-binder. Comparison with recently available benchmarking studies indicated that, the prediction accuracy of our method for many of the class I and class II MHC alleles was comparable to the sequence based methods, even if it does not use any experimental data for training. It is also encouraging to note that the ranks of true binding peptides could further be improved, when high scoring peptides obtained from pair potential were re-ranked using all atom forcefield and MM/PBSA method.

The GD box: a widespread noncontiguous supersecondary structural element

Identification and characterization of recurrent supersecondary structural elements is central to understanding the rules governing protein tertiary structure. Here, we describe the GD box, a widespread noncontiguous supersecondary element, which we initially found in a group of topologically distinct but homologous beta-barrels--the cradle-loop barrels. The GD box is similar both in sequence and structure and comprises two short unpaired beta-strands connected by an orthogonal type-II beta-turn and a noncontiguous beta-strand forming hydrogen bonds with the beta-turn. Using structure-based analysis, we have detected 518 instances of the GD box in a nonredundant subset of the SCOP database comprising 3771 domains. Apart from the cradle-loop barrels, this motif is also found in a diverse set of nonhomologous folds including other topologically related beta-barrels. Since nonlocal interactions are fundamental in the formation of protein structure, systematic identification and characterization of other noncontiguous supersecondary structural elements is likely to prove valuable to protein structure modeling, validation, and prediction.