Workshop on OR in Computational Biology, Bioinformatics, and Medicine

 

Invited Talks

Title:         Advances and Challenges in De Novo Protein Design Authors:   Christodoulos A. Floudas                    Princeton University, NJ, USA Abstract: The primary objective in de novo protein design is to determine the amino acid sequences which are compatible with existing or postulated template backbone structures that may be rigid or flexible. The de novo protein design problem is of fundamental importance since it addresses the mapping of the space of amino acid sequences to known protein folds or postulated/putative protein folds. It is also of significant practical importance since it can lead to the improved design of inhibitors, design of novel sequences with better stability, design of catalytic sites of enzymes, and drug discovery. The first part of this lecture will provide a motivation for the de novo protein design problem, a definition of the flexible backbone template structures, and an overview of the advances and limitations. The second part will introduce a novel two-stage approach which takes into account explicitly the flexibility of the templates. The first stage addresses the in silico sequence selection problem through two key contributions: (a) the development of a distance-dependent Ca-Ca and side chain centroid-centroid distance dependent force fields; and (b) a rigorous quadratic assignment-like formulation for the prediction of a rank-ordered list of sequences with novel mutations. The second stage addresses the fold specificity problem by performing structure prediction calculations using atomistic level force fields. Two alternative approaches will be presented for the generation of ensembles of protein conformations: (i) the first principles protein structure prediction approach, Astro-Fold, and (ii) an approach motivated by an established NMR structure refinement protocol. Based on the ensembles of protein structures generated, the probabilities of each predicted sequence to fold specifically to the flexible templates are calculated. The theoretical prediction results for several peptides and proteins that include variants of Compstatin, human beta defensin-2, C3a, and gp41 for HIV-1 will be presented. Comparisons with experimental findings will also be discussed.

Title:        Protein Dynamics and Allostery: Insights from Network Models Author:    Ivet Bahar                   University of Pittsburgh, PA, USA Abstract: Elastic network models have been widely used in recent years for describing protein dynamics. The biomolecular structure is represented therein by a network of beads and springs to examine the collective dynamics of the system of harmonic oscillators. Many studies have shown that the most cooperative modes of motions accessible to biomolecular systems are relevant to their functional motions. Motivated by the success of elastic network models, we have exploited the utility of graph theory and spectral graph methods for exploring the pathways of communication in allosteric proteins. To this aim a Markovian diffusion of information across the structure is assumed and the hitting times between residue pairs are examined. Examination of the information processing characteristics of biomolecular structures reveal the efficient communication tendencies of catalytic and conserved residues, as well as the important role of tertiary contacts between secondary structural elements.

Title:           Streaming facts from scientific publications to the scientist and new frontiers in publishing Author(s):  Dietrich Rebholz-Schuhmann                      EBI, Cambridge, UK Abstract: Scientific literature is increasingly available in electronic form and early on after its acceptance for publication. Techniques to analyse the literature for contained facts are then applied to deliver individual facts directly to the scientist. This leads to the integration of the scientific literature into the infrastructure of existing IT data resources. This talk will explain how scientists in the biomedical domain profit from an infrastructure consisting of services for information extraction. All services automatically process the documents and interlink them with bioinformatics data resources. In addition they can be integrated into external IT solutions to directly couple experimental results with annotations from the scientific literature and to new solutions to support scientists in their pre-submission publication preparation process.