Accepted Posters
Category 'C'- Chemical and Pharmaceutical Informatics' |
Poster C01 |
Molwind – Exploring Molecule Spaces through Geospatial Browsing |
Oliver Karch- Merck KGaA |
Christian Herhaus (Merck KGaA, Bio- and Chemoinformatics); |
Short Abstract: High-throughput-techniques used by "omics" research produce large sets of molecular entities which often need to be explored interactively to allow simultaneous inspection of high-dimensional properties. Here we propose Molwind (http://www.molwind.org) to visually explore molecule spaces. It builds on NASA's geospatial browser World Wind and offers unique features for intuitive navigation and view-dependent level-of-detail rendering |
Long Abstract:Click Here |
|
Poster C02 |
In silico screening for Phospholipase A2 of Apis Mellifera with HitFinder Library |
Daniel Jorge- University of Sao Paulo |
Vinicius B. da Silva (University of São Paulo, School of Science of Pharmacy of Ribeirao Preto); Carlos H. T. P. Silva (University of São Paulo, School of Science of Pharmacy of Ribeirao Preto); Andreimar M. Soares (University of São Paulo, School of Science of Pharmacy of Ribeirao Preto); Silvana Giuliatti (University of São Paulo, Department of Genetics); |
Short Abstract: Phospholipases A2 (PLA2s) are enzymes that catalyze the hydrolysis of the sn-2 acyl bond of glycerophospholipids. The aim of this project was predict potential inhibitors against bvPLA2. The best score compost has been named Molecule 8657. Fifty proposals were selected and also should be evaluated against bvPLA2. |
Long Abstract:Click Here |
|
Poster C03 |
In Silico Molecular Drug Design of Platelet Factor 4 Antagonists |
John Rux- In Silico Molecular, LLC. |
Ann Rux (University of Pennsylvania, Department of Pathology and Laboratory Medicine); Jillian Hinds (University of Pennsylvania, Department of Pathology and Laboratory Medicine); Bruce Sachais (University of Pennsylvania, Department of Pathology and Laboratory Medicine); |
Short Abstract: Our aim is to develop a drug to treat heparin-induced thrombocytopenia and thrombosis (HITT), a serious complication of heparin therapy. Using computational chemistry we have identified several small molecules that antagonize Platelet Factor 4 tetramerization in vitro, demonstrating proof of principle for our computational approach and target binding site. |
Long Abstract:Click Here |
|
Poster C04 |
Novel Fragment-based Quantitative Activity Prediction Method for Cannabinoid Drug Design |
Kyaw Zeyar Myint- University of Pittsburgh, Carnegie-Mellon University |
Chao Ma (University of Pittsburgh/Carnegie-Mellon University, Computational Biology); Xiang-Qun Xie (University of Pittsburgh/Carnegie-Mellon University, Computational Biology/Pharmacy); Lirong Wang (University of Pittsburgh, Pharmacy); |
Short Abstract: We introduced a novel computational method to predict biological activities of a family of cannabinoid ligands by correlating known ligand activities with molecular fragments using multivariate linear regression. With the introduction of the fragment-similarity concept using fingerprints and BCUT matrix calculations, the method outperformed the original Free-Wilson method by 61%. |
Long Abstract:Click Here |
|
Poster C05 |
Designing and implementing chemoinformatic approaches in TDR Targets Database: linking genes to chemical compounds in tropical disease causing pathogens |
María Magariños- Universidad de San Martín |
John Overington (EBML Outstation, Hinxton, Cambridge, UK, European Bioinformatics Institute); Santiago Carmona (Universidad de San Martín, Instituto deinvestigaciones bioquímicas); Fernán Agüero (Universidad de San Martín, Instituto deinvestigaciones bioquímicas); Dhanasekaran Shanmugam (University of Pennsylvania, Philadelphia, PA (USA), Department of biology); David Roos (University of Pennsylvania, Philadelphia, PA (USA), Department of biology); Maria Doyle (University of Melbourne, Victoria (Australia), Molecular science and biotechnology institute); Stuart Ralph (University of Melbourne, Victoria (Australia), Molecular science and biotechnology institute); Greg Crowther (University of Washington, Seattle WA (USA), Division of allergy and infectious deseases); Wes Van Voorhis (University of Washington, Seattle WA (USA), Division of allergy and infectious deseases); Christiane Hertz-Fowler (Wellcome Trust Sanger Institute, Hinxton, Cambridge (UK), Parasite genomics); Matt Berriman (Wellcome Trust Sanger Institute, Hinxton, Cambridge (UK), Parasite genomics); Solomon Nwaka (WHO/TDR, Geneva (Switzerland), Special Programme for Research and Training in Tropical Diseases (TDR)); |
Short Abstract: TDR Targets (tdrtargets.org) is a database that associates gene information from human pathogens with genomic and functional information. In this work, information about 504,020 drug-like compounds was integrated into TDR Targets, where users can search the database to get compound information and relate compounds with pathogen genes. |
Long Abstract:Click Here |
|
Poster C06 |
Improved Visualization of Pathway Prediction |
Junfeng Gao- University of Minnesota |
Lynda Ellis (University of Minnesota, Laboratory Medicine and Pathology); Larry Wackett (University of Minnesota, Biochemistry, Molecular Biology, and Biophysics); |
Short Abstract: The UM Pathway Prediction System (UM-PPS, http://umbbd.msi.umn.edu/predict/) is a rule-base system that predicts microbial catabolism of organic compounds. The improved visualization has the capability to show common intermediates and cleavage products in a multi-step prediction. Users can now view prediction alternatives much more easily. |
Long Abstract:Click Here |
|
Poster C07 |
Chemical structure search in PubChem |
Jie Chen- National Center for Biotechnology Information |
Vahan Simonyan (National Center for Biotechnology Information, Computational Biology Branch); Wolf-D Ihlenfeldt (Xemistry GmbH, N/A); Paul Thiessen (National Center for Biotechnology Information, Computational Biology Branch); Evan Bolton (National Center for Biotechnology Information, Computational Biology Branch); Stephen Bryant (National Center for Biotechnology Information, Computational Biology Branch); |
Short Abstract: Chemical structure search is a critical component in PubChem. It provides a powerful and easily accessible gateway with efficient large-scale data mining ability to explore small molecules sharing structural or physical similarities. |
Long Abstract:Click Here |
|
Poster C08 |
Association of feature gene expression with structural fingerprints of chemical compounds |
Pei Hao- Shanghai Institutes for Biological Sciences, Chinese Academy of Science |
Yun Li (Shanghai Institutes for Biological Sciences, Chinese Academy of Science, Key Laboratory of System Biology); Xuan Li (Shanghai Institutes for Biological Sciences, Chinese Academy of Science, Institute of Plant Physiology &Ecology); Yixue Li (Shanghai Institutes for Biological Sciences, Chinese Academy of Science, Key Laboratory of System Biology); |
Short Abstract: By combining tanimoto coefficients-based chemical structure classification with chemical intervention related microarray expression data, we presented a novel method which aims to find feature genes that reflect the functional differences of chemicals caused by the structural differences, so as to build relationships between chemical structure and their inner functions. |
Long Abstract:Click Here |
|
Poster C09 |
Novel Ligand Classification Algorithm and Application on Modeling Functionality for 5HT1A GPCR Ligands |
Chao Ma- University of Pittsburgh |
Lirong Wang (University of Pittsburgh, School of Pharmacy); Kyaw Zeyar Myint (University of Pittsburgh, Computational Biology); Xiangqun Xie (University of Pittsburgh, School of Pharmacy); |
Short Abstract: We have developed a novel algorithm, Ligand Classifier of Adaptively Boosting Ensemble Decision Stump (LiCABEDS), and validated through ligand functionality predictions, showing average 90% accuracy in classifying agonists/antagonists of 5HT1A GPCR ligands. Implemented in user-friendly interface, LiCABEDS also demonstrates its robustness, flexibility and interpretability in modeling quantitative structure-property relationship. |
Long Abstract:Click Here |
|
Poster C10 |
Kinase Inhibition Modeling by Proteochemometrics and Docking Analysis |
Akinori Sarai- Kyushu Institute of Technology |
Michael Fernández (Kyushu Institute of Technology, Department of Bioscience and Bioinformatics); Shandar Ahmad (National Institute of Biomedical Innovation, Bioinformatics Division); |
Short Abstract: Selective kinase inhibitors constitute an alternative clinical strategy for cancer treatment. We applied docking and topological autocorrelation's approach to model our in-house manually-curated and annotated kinase-ligand interaction data in ProLINT database. Support Vector Machines correctly classified 83% of the complexes to be stable or unstable. The predictor is available online. |
Long Abstract:Click Here |
|
Poster C11 |
A probabilistic method for identifying transient binding pockets on protein surfaces |
Paul Ashford- Birkbeck, University of London |
Irilenia Nobeli (Institute of Structural and Molecular Biology, Birkbeck, University of London, Malet Street, London, WC1E 7HX, Biological Sciences); Alexander Alex (Pfizer, Ramsgate Road, Sandwich CT13 9NJ, Global Research and Development); David Moss (Institute of Structural and Molecular Biology, Birkbeck, University of London, Malet Street, London, WC1E 7HX, Biological Sciences); Mark Williams (Institute of Structural and Molecular Biology, Birkbeck, University of London, Malet Street, London, WC1E 7HX, Biological Sciences); |
Short Abstract: We present a method for investigating the occurrence of surface binding pockets across sets of protein structures, such as those obtained from Essential Dynamics, homologous structures or NMR ensembles. These residue-based probabilities of bordering a pocket could be useful guides to identifying druggable sites for inhibition of protein-protein interactions. |
Long Abstract:Click Here |
|
Poster C12 |
PrOCoil - Advances in predicting two- and three-stranded coiled coils |
Carsten Mahrenholz- Charité Medical School, Berlin |
Ingrid G. Abfalter (Johannes Kepler University, Institute of Bioinformatics); Ulrich Bodenhofer (Johannes Kepler University, Institute of Bioinformatics); Rudolf Volkmer (Charité Medical School, Institute of Medical Immunology); Sepp Hochreiter (Johannes Kepler University, Institute of Bioinformatics); |
Short Abstract: A complex network of amino acid dependencies is the driving force of coiled coil stoichiometry. Our online-tool PrOCoil, classifies coiled coils with an outstanding accuracy of 86% and is also able to visualize the contribution of each individual amino acid to the overall oligomeric tendency of a given coiled coil sequence. |
Long Abstract:Click Here |
|
Poster C13 |
HALO384: A Halo-Based Potency Prediction Algorithm for High-Throughput Detection of Antimicrobial Agents |
Marcos Woehrmann- UC Santa Cruz |
Nadine Gassner (UC Santa Cruz, Chemistry); Walter Bray (UC Santa Cruz, Chemistry); Josh Stuart (UC Santa Cruz, BME); Scott Lokey (UC Santa Cruz, Chemistry); |
Short Abstract: HALO384 is a high-throughput (HT) agar-based halo assay that uses a pattern recognition algorithm to identify halo-like shapes in plate reader optical density (OD) measurements to allow for rapid screening of chemical libraries for bio-activity in microorganisms such as yeast and bacteria. |
Long Abstract:Click Here |
|
Poster C14 |
Systematic drug target discovery via chemical and genetic interaction profiles |
Frederick Phillip Roth- Harvard Medical School |
Hon Nian Chua (Harvard Medical School, Department of Biological Chemistry and Molecular Pharmacology); |
Short Abstract: A variety of genome-scale studies have revealed the interplay between genes and drugs, and pointed to mechanisms of drug action. Haploinsufficiency profiling (HIP) can often reveal drug targets directly, while Homozygous Profiling (HOP) tends to reveal genes that compensate for loss of the drug target's activity. Ge-netic interaction profiles can be combined with HOP to reveal drug targets (HOP-GI). Although there are anecdotal examples of using each approach to uncover drug targets, there has not been a quantitative assessment of the success against a 'gold standard' set of known drug/target relationships for any of these methods, either individually or taken together. Here we describe the systematic evaluation of genetic approaches to reveal drug targets. We integrated 538 ge-nome-scale chemical-genomic profiles data with genetic interaction profiles cov-ering 75% of all genes in S. cerevisiae. Using known drug/target relationships assembled from the literature, we found that HIP alone can detect 23% of known drug targets (48% of those examined by HIP). An integrated approach combining HIP and HOP-GI detects 67% of known targets examined by HIP and HOP-GI. Our results provide new drug target predictions and support the continued appli-cation of large-scale studies of genetic and chemical-genomic profiles to system-atically reveal mechanisms of drug action. |
Long Abstract:Click Here |
|
Poster C15 |
Discovering Synergistic Drug Pairs Using Genetic Interactions |
Frederick Roth- Harvard Medical School |
No additional authors |
Short Abstract: Pathogens and tumors that respond poorly to single drugs may be amenable to a 'synergistic drug pair,' whereby two drugs administered together are collectively more potent than the independent effects of each individual drug. Such drug interactions are analogous to synergistic genetic interactions: Two genes are defined to have a synergistic genetic interaction when the impairment of both genes results in slower than expected growth rate. Assuming a drug's action on a target protein is similar to the hypomorphic or null allele of the gene coding the target, we conjectured that two drugs will be synergistic when they target proteins encoded by two synergistic genes. An analysis of the existing drug-target relationship and genetic interaction datasets for S. cerevisiae revealed five drug pairs that target synergistic genes. One of these pairs was dropped from the analysis since it included a membrane disrupting agent, which is promiscuously synergistic with a wide variety of drugs. We conducted more than 400,000 cell density measurements in various drug combinations and concentrations to test our conjecture. We found drug synergy for two of the four drug pairs targeting synergistic genes, but found no significant synergy amongst any of the other seventeen drug pairs tested. This indicates a significant overrepresentation of synergistic drug pairs among drug pairs that target synergistic and demonstrates that genetic interactions can serve as a useful predictor for discovering drug synergy. |
Long Abstract:Click Here |
|
Accepted Posters
↑ TOP
|