Accepted Posters

Category 'W'- Structure and Function Prediction'
Poster W01
The a-galactosidase type A gene aglA from Aspergillus niger encodes a fully functional a-N-acetylgalactosaminidase
Natallia Kulik- University of South Bohemia
Lenka Weignerová (Academy of Sciences of the Czech Republic, Center for Biocatalysis and Biotransformation, Institute of Microbiology); Tomáš Filipi (Academy of Sciences of the Czech Republic, Center for Biocatalysis and Biotransformation, Institute of Microbiology); Petr Novák (Academy of Sciences of the Czech Republic, Institute of Microbiology, Laboratory of Molecular Structure Characterization); Petr Pompach (Academy of Sciences of the Czech Republic, Institute of Microbiology, Laboratory of Molecular Structure Characterization); Hynek Mrázek (Charles University, Department of Biochemistry, Faculty of Science); Karel Bezouška (Academy of Sciences of the Czech Republic, Institute of Microbiology, Laboratory of Molecular Structure Characterization); Kristýna Slámová (Academy of Sciences of the Czech Republic, Center for Biocatalysis and Biotransformation, Institute of Microbiology); Vladimír K?en (Academy of Sciences of the Czech Republic, Center for Biocatalysis and Biotransformation, Institute of Microbiology); Rüdiger Ettrich (Academy of Sciences of the Czech Rep, Center for Biocatalysis and Biotransformation, Institute of Systems Biology and Ecology);
Short Abstract: Molecular dynamics simulations, homology modelling and substrate docking are applied to investigate structural differences determining the selectivity in ?-galactosidase variants A and B from A.niger, which lead to the assignment of the experimentally characterized putative ?-N-Acetylgalactosaminidase to gene variant A of ?-galactosidase from A.niger.
Long Abstract:Click Here

Poster W02
DescFold:A web server for protein fold recognition
Yan Xiang- China Agricultural University
No additional authors
Short Abstract: In this work, a machine learning-based method called DescFold was established by using Support Vector Machines to combine different sequence and structural information. Meanwhile, we also set up a user friendly web server that implements the method.
Long Abstract:Click Here

Poster W03
Ab initio Prediction of E5 Protein Structure from Cervical Human Papillomavirus (HPV)
Nilson Nicolau Junior- University of São Paulo
Cristina Junta ( University of São Paulo, Department of Genetics); Silvana Giuliatti ( University of São Paulo, Department of Genetics);
Short Abstract: Invasive cervical carcinoma arises in persistent infection by HPV. Due the importance of the oncoprotein E5 in the carcinogenesis, obtaining a 3D structure model of it may facilitate studies involving protein-protein and inhibitor-protein docking experiments. This present work predicted and validated E5 protein structures from cervical cancer high risk HPVs.
Long Abstract:Click Here

Poster W04
Advanced TOPS+ Comparison Method for Enhanced TOPS models
Mallika Veeramalai- Sanford-Burnham Medical Research Insititute
David Gilbert (Brunel University, School of Information Systems, Computing and Mathematics); Gabriel Valiente (Technical University of Catalonia, Algorithms, Bioinformatics, Complexity and formal methods research group);
Short Abstract: Although methods based on highly abstract descriptions of protein structures, such as VAST and TOPS, can perform very fast protein structure comparison, the results can lack a high degree of biological significance. Here we present our advanced TOPS+ comparison method for enhanced TOPS models.
Long Abstract:Click Here

Poster W05
Sequence-to-structure prediction errors can reveal conformational flexibility in helical membrane proteins
Shandar Ahmad- National Institute of Biomedical Innovation
Hemjit Singh (National Institute of Biomedical Innovation, Bioinformatics); Yogesh Paudel (National Institute of Biomedical Innovation, Bioinformatics); Takaharu Mori (RIKEN, Advaced Science Institute); Yuji Sugita (RIKEN, Advaced Science Institute); Kenji Mizuguchi (National Institute of Biomedical Innovation, Bioinformatics);
Short Abstract: In this work, we show that for membrane proteins with known structure, accurate estimates of flexible residues can be made by comparing their observed structure features with those predicted from sequence information. Together, normal modes analysis and these prediction errors can predict flexible at a correlation coefficient ~0.91.
Long Abstract:Click Here

Poster W06
Developing an integrated, knowledge-based, protein-protein docking scoring function
Jawon Song- Cambridge University
Tom Blundell (Cambridge University, Biochemistry);
Short Abstract: Integration of different protein-protein interaction parameters into protein-protein docking scoring functions gives an improved performance compared to more conventional approaches. The inclusion of conservation scores in scoring functions along with energetic terms and residue pair potentials is described in this study.
Long Abstract:Click Here

Poster W07
Accurate Antibody Complementarity Determining Region Loop Structure Prediction Using Contact Profeils
Yoonjoo Choi- Oxford University
Charlotte Deane (Oxford University, Statistics);
Short Abstract: FREAD-C3 is a new loop prediction method that uses contact information in order to accurately predict antibody CDRs. It outperforms other available methods giving predictions for CDR-H3 loops to within 1.5? RMSD on average (coverage 83%) on a test set of 162 loops.
Long Abstract:Click Here

Poster W08
Protein secondary structure appears to be robust under in silico evolution while protein disorder appears not to be.
Christian Schaefer- Columbia University
Avner Schlessinger (University of California , Bioengineering and Therapeutic Sciences); Burkhard Rost (Columbia University, Biochemistry and Molecular Biophysics);
Short Abstract: We here study the robustness of predicted protein secondary structure as well as disorder under in silico mutation. We find that secondary structure appears to be robust in terms of content and length while long disorder appears to vanish after a few mutation steps.
Long Abstract:Click Here

Poster W09
Functional dynamics of NhaA predict motion implicated in alternating access and pH-induced activation
Maya Schushan- Tel-Aviv University
Etana Padan (Hebrew University, Biological Chemistry); Turkan Haliloglu (Bo?aziçi University, Chemical Engineering); Nir Ben-Tal (Tel-Aviv University, Biochemistry and molecular biology);
Short Abstract: We investigated the dynamics of NhaA, a bacterial pH regulated Na+/H+ antiporter, using elastic network models. We link the most cooperative (first) mode to the ion transport mechanism and alternate accessing. The second predicts a possible interaction between the pH sensor and the functional site, ascribed to pH regulation.
Long Abstract:Click Here

Poster W10
ConQuass: using evolutionary conservation for quality assessment of protein model structures
Matan Kalman- Tel Aviv University
Nir Ben-Tal (Tel Aviv University, Department of Biochemistry and Molecular Biology);
Short Abstract: We present ConQuass, a program that allows quality assessment of protein model structures based only on the consistency between the model structure and the protein's evolutionary conservation pattern. We show that the score it assigns a model correlates with the similarity of the model to the native structure.
Long Abstract:Click Here

Poster W11
Prediction of dinucleotide-specific RNA-binding sites in proteins using sequence and evolutionary information
Michael Fernandez- Kyushu Institute of Technology
Yutaro Kumagai (Osaka University, Immunology Frontier Research Center (IFReC)); Dayron Stanley (Osaka University, Immunology Frontier Research Center (IFReC)); Akinori Sarai (Kyushu Institute of Technology, Bioscience and Bioinformatics); Kenji Mizuguchi (Bioinformatics Laboratory, National Institute of Biomedical Innovation); Ahmad Shandar (Bioinformatics Laboratory, National Institute of Biomedical Innovation);
Short Abstract: Abstract
Regulation of gene expression, protein synthesis and replication involve RNA–protein interactions. We have developed a neural network model using sequence and evolutionary information that recognizes dinucleotide-specific RNA-binding sites in proteins with accuracies of ~70-80%. Furthermore, predicted protein affinity profiles for RNA sequences positively correlated with the experimental binding data.
Long Abstract:Click Here

Poster W12
Automated All-Atom RNA Tertiary Structure Prediction
Matthew Seetin- University of Rochester Medical Center
Matthew Seetin (University of Rochester Medical Center, Biochemistry and Biophysics); David Mathews (University of Rochester Medical Center, Biochemistry and Biophysics);
Short Abstract: A novel protocol employing simulated annealing and steered molecular dynamics for all-atom RNA tertiary structure prediction is presented. The restraints are derived from secondary structure, coaxial stacking predictions for helices in multi-way gunctions, co-variation analysis, and, when available, cross-linking data. It was tested on 4 RNAs.
Long Abstract:Click Here

Poster W13
Multiple protein docking prediction based on genetic algorithms and physics-based scoring
Juan Esquivel-Rodriguez- Purdue University
Daisuke Kihara (Purdue University, Biological Sciences/Computer Science);
Short Abstract: The current work proposes using graph-based genetic algorithms to address the multiple protein docking problem. Combining a shape-oriented search first, using our pairwise docking program LZerD, followed by a genetic algorithm based search procedure, driven by a physics-based scoring function, allows us to efficiently search the conformational space
Long Abstract:Click Here

Poster W14
MEDELLER: Homology-Based Coordinate Generation for Membrane Proteins
Sebastian Kelm- University of Oxford
Jiye Shi (UCB Celltech, ROC); Charlotte Deane (University of Oxford, Dept of Statistics);
Short Abstract: MEDELLER is a homology-based coordinate generation method for membrane proteins. Using membrane protein-specific information and accurate loop modelling it outperform the popular method Modeller on all test sets by an average of 0.48 Angstroms RMSD.
Long Abstract:Click Here

Poster W15
Evolution of apparently inactive active sites in the HexxH metalloproteases
Teresa Szczepinska- Nencki Institute of Experimental Biology
Anna Lenart (Nencki Institute of Experimental Biology, Department of Molecular and Cellular Neurobiology); Krzysztof Paw?owski (Nencki Institute of Experimental Biology, Department of Molecular and Cellular Neurobiology);
Short Abstract: Conserved non-random substitutions in HExxH motif metalloprotease active sites are analysed. In CLCAs, putative metalloproteases, "broken-down" enzyme is phylogenetically common. Structure modelling of variants and comparisons to site databases allow hypotheses on functional roles.
Long Abstract:Click Here

Poster W16
A machine learning application for the high-performance prediction of functional sites in proteins of unknown function
Srinivas Somarowthu- Northeastern University
Huyuan Yang (Northeastern University, Chemistry and Chemical Biology); Yujing Wang (Northeastern University, Informatics); Andrew Michaelson (Northeastern University, Biology); Mary Jo Ondrechen (Northeastern University, Chemistry and Chemical Biology);
Short Abstract: Many protein structures are annotated as 'hypothetical' or 'unknown function'; computational methods are needed to predict functional information. We present a new machine learning method POOL with new input features for functional residue prediction. Improved performance is achieved using THEMATICS electrostatics data and sequence-based phylogenetic information from INTREPID.
Long Abstract:Click Here

Poster W17
CompaRNA: a server for continuous benchmarking of automated methods for RNA structure prediction.
Tomasz Puton- Faculty of Biology, Adam Mickiewicz University
Kristian Rother (International Institute of Molecular and Cell Biology in Warsaw, Laboratory of Bioinformatics and Protein Engineering); ?ukasz Koz?owski (International Institute of Molecular and Cell Biology in Warsaw, Laboratory of Bioinformatics and Protein Engineering); Ewa Tkalin?ska (International Institute of Molecular and Cell Biology in Warsaw, Laboratory of Bioinformatics and Protein Engineering); Janusz M. Bujnicki (International Institute of Molecular and Cell Biology in Warsaw, Laboratory of Bioinformatics and Protein Engineering);
Short Abstract: The CompaRNA web server provides a continuous benchmark of automated methods for RNA secondary structure prediction. The aim of CompaRNA is to assess the state of the art in RNA structure prediction, provide a detailed picture of what is possible with the available tools, where progress is being made and what major problems remain.
Long Abstract:Click Here

Poster W18
Functional Annotation of a Structural Genomics Protein from the ClpP/crotonase Superfamily using Local Structure Analysis
Pengcheng Yin- Northeastern University
Mary Jo Ondrechen (Northeastern University, Chemistry and Chemical Biology);
Short Abstract: A new computational methodology is used to annotate the function of a structural genomics protein that was annotated as a putative enoyl-CoA hydratase. We show that it is more likely a CoA-dependent dehalogenase based on 3D structural alignment, analyzed in the local region of the computationally predicted active site.
Long Abstract:Click Here

Poster W19
Parametric maximum expected accurate RNA structure
Feng LOU- Paris Sud 11 University
Peter CLOTE (Boston College, Biology);
Short Abstract: Using McCaskill's algorithm and dynamic programming, we present a new algorithm to compute, for each k, the Maximum Expected Accurate RNA secondary structure, MEA(k), having base pair distance k from an initial structure. Our motivation is to improve the prediction of riboswitches in long 5' untranslated regions of mRNA.
Long Abstract:Click Here

Poster W20
Modeling protein-peptide interactions using protein fragments: fitting the pieces?
Peter Vanhee- Switch Laboratory, VIB
Peter Vanhee (Switch Laboratory VIB, Department of Biotechology, VUB, Brussels); Frederic Rousseau (Switch Laboratory VIB, Department of Biotechology, VUB, Brussels); Joost Schymkowitz (Switch Laboratory VIB, Department of Biotechology, VUB, Brussels); Lies Baeten (Switch Laboratory VIB, Department of Biotechology, VUB, Brussels); Erik Verschueren (Center for Genomic Regulation, EMBL/CRG Systems Biology Unit, Barcelona); Francois Stricher (Center for Genomic Regulation, EMBL/CRG Systems Biology Unit, Barcelona); Luis Serrano (Center for Genomic Regulation, EMBL/CRG Systems Biology Unit, Barcelona);
Short Abstract: We compared the modes of interaction between protein-peptide interfaces and those observed within monomer proteins and found surprisingly little difference. Over 65% of 731 protein-peptide interfaces could be reconstructed within 1 Å RMSD using solely fragment interactions, suggesting that our fragment interaction approach might provide an alternative to homology modeling.
Long Abstract:Click Here

Poster W21
PETcofold: Prediction of structurally conserved RNA-RNA interactions
Andreas Richter- University of Freiburg
Stefan E. Seemann (University of Copenhagen, Center for non-coding RNA in Technology and Health, IBHV); Tanja Gesell (University of Vienna, Center for Integrative Bioinformatics Vienna (CIBIV), Max F. Perutz Laboratories (MFPL)); Jan Gorodkin (University of Copenhagen, Center for non-coding RNA in Technology and Health, IBHV); Rolf Backofen (University of Freiburg, Bioinformatics Group);
Short Abstract: We present PETcofold, a new approach for the prediction of RNA-RNA interactions using two multiple alignments of RNA sequences. PETcofold uses covariance information in both intramolecular and intermolecular base pairs to predict the joint secondary structure of two RNAs including their interaction. PETcofold was evaluated on eukaryotic and bacterial interactions.
Long Abstract:Click Here

Poster W22
Predicting ligand binding residues across the Human Genome
Gonzalo Lopez- Cnio
Gonzalo Lopez (CNIO, SBBP); Michael Tress (CNIO, SBBP); Alfonso Valencia (CNIO, SBBP); Jose Manuel Rodriguez (CNIO, SBBP);
Short Abstract: We present the new developments of firestar, an expert system for
predicting functional residues based on protein structures. Firestar has been automatized and a new web interface has been made available.
We have applied firestar in high throughput mode to predict functional
residues and to cataloge ligand binding sites across the human genome.
Long Abstract:Click Here

Poster W23
Identification of functional subclasses in the ribulose-phosphate binding barrel superfamily using computational tools
Joslynn Lee- Northeastern University
Joslynn Lee (Northeastern University, Chemistry and Chemical Biology); Srinivas Somarowthu (Northeastern University, Chemistry and Chemical Biology); Mary Jo Ondrechen (Northeastern University, Chemistry and Chemical Biology);
Short Abstract: Prediction of the function of a protein from its three-dimensional structure is important. A computational method is described here for the identification of function of proteins within the RPBB superfamily. Functional assignments within this superfamily are based on a local structural analysis at the known or predicted interaction sites.
Long Abstract:Click Here

Poster W24
High accuracy on-lattice side chain models of PDB protein structures
Martin Mann- University of Freiburg
Rhodri Saunders (University of Oxford, Department of Statistics); Cameron Smith (University of Freiburg, Bioinformatics Group); Rolf Backofen (University of Freiburg, Bioinformatics Group); Charlotte M. Deane (University of Oxford, Department of Statistics);
Short Abstract: Lattice proteins are a common abstraction to study protein structures and related topics. LatFit is a new tool to derive accurate sidechain lattice models of PDB protein structures using a distance RMSD-optimisation procedure. The web interface enables direct use and we show its superior performances compared to results from literature.
Long Abstract:Click Here

Poster W25
Rapid Estimation of RNA Kinetics
Andy Lorenz- Boston College
Peter Clote (Boston College, Biology);
Short Abstract: Kinetics of RNA folding is known to play critical regulatory roles in many cellular processes. We describe a novel algorithm for the partition function of locally optimal structures. By computing MFPT along locally optimal structures, we obtain fast,
accurate approximations of folding kinetics, an ideal tool
for RNA design.
Long Abstract:Click Here

Poster W26
MUFOLD: A new solution for protein 3D structure prediction
Jingfen Zhang- University of missouri
Dong Xu (University of Missouri, Computer Science); Qingguo Wang (University of Missouri, Computer Science); Zhiquan He (University of Missouri, Computer Science); Yi Shang (University of Missouri, Computer Science); Bogdan Barz (University of Missouri, Physics and Astronomy); Ioan Kosztin (University of Missouri, Physics and Astronomy);
Short Abstract: One important tool to bridge the gap between existing protein sequences and structures is computational prediction. We have developed a new system 'MUFOLD', which is much fast than current popular tools such as 'Rosetta' and 'Modeller' while keeps the similar prediction accuracy.
Long Abstract:Click Here

Poster W27
RNA Structural-BLAST: Unleashing the Power of RNA Structure by Using RNA Fingerprint
KEJIE LI- Purdue University
Michael Gribskov (Purdue University, Department of Biological Sciences);
Short Abstract: Based on the XIOS framework, we present a new RNA topology based indexing and structural comparison tool: RNA Structural-BLAST. It can index all the RNA structures, and produce unique RNA structural spectral fingerprints for fast structural comparison purpose. It also provides fast database searching, matching and classification of RNA.
Long Abstract:Click Here

Poster W28
Evaluation of the capabilities of the Consensus of Prediction (CoP) Approach for the Prediction of Catalytic Residues Implemented in the ENDURANCE
Natalia Petrova- University Of Delaware
No additional authors
Short Abstract: An accurate automatic prediction of the functional residues plays an important role in the post genome-sequencing era for a large-scale analysis/annotation of protein function. Here, we demonstrate a predictive power of the automated tool – ENDURANCE – that is capable of collecting all relevant information; analysis and visualization of the prediction results.
Long Abstract:Click Here

Poster W29
Finding Conserved Topology in RNA Structure
Reazur Rahman- Purdue University
No additional authors
Short Abstract: Discovering conserved structural motifs in a set of RNA structures is likely to have functional and structural significance. We develop XIOS RNA structure matching tool to find conserved structural topology in different sets of RNA structures, including predicted suboptimal free energy structures.
Long Abstract:Click Here

Poster W30
Gapped GDT_TS – A score for detecting variations in structural similarity
Sheenal Srivastava- Macquarie University
Graham Wood (Macquarie University, Statistics);
Short Abstract: A Gapped GDT_TS score is developed to measure areas of strong and weak fit within a superposition, providing a diagnostic tool to assess predictions and thus prediction methods.
Long Abstract:Click Here

Poster W31
GeneMANIA: Intelligent gene function prediction and interactive network visualization
Gary Bader- University of Toronto
Quaid Morris (University of Toronto, Donnelly Centre for Cellular and BioMolecular Research);
Short Abstract: GeneMANIA ( generates hypotheses about gene function by querying large?scale publicly available genomics and proteomics datasets. GeneMANIA is unique; users enter a gene list, and GeneMANIA extends the list and generates a composite network in seconds. Users interact with the composite network, exploring in detail the predicted genes.
Long Abstract:Click Here

Poster W32
A network based study of GPCRs
AJ Venkatakrishnan- MRC Lab of Molecular Biology / University of Cambridge
Gebhard Schertler (Paul Scherrer Institut, Laboratory of Biomolecular Research, BMR); Madan Babu Mohan (MRC Lab of Molecular Biology, Lab of Molecular Biology);
Short Abstract: G-protein coupled receptors (GPCRs) are transmembrane signalling proteins that are central to health and disease. We aim to elucidate the process of signal transduction at the molecular level for Rhodopsin, a representative member of Class A GPCRs, by studying the structural intermediates using a netwok based approach.
Long Abstract:Click Here

Poster W33
RNA structural segmentation
Peter Clote- Boston College
Ivan Dotu (Dynadec , Combinatorial Optimization); William A. Lorenz (Boston College, Biology); Pascal Van Hentenryck (Brown University, Computer Science);
Short Abstract: We describe several dynamic programming segmentation algorithms to segment
RNA secondary and tertiary structure into domains. Our base
fitness functions is variously defined using
(i) base pairing probabilities in the Boltzmann low energy ensemble of
structures, (ii) contact maps inferred from 3-dimensional structures,
(iii) Voronoi tessellation computed from
3-dimensional structures.
Long Abstract:Click Here

Poster W34
New sequence-based and structure-based methods for predicting class II MHC epitopes
Andrew Bordner- Mayo Clinic
Andrew Bordner (Mayo Clinic, Department of Molecular Pharmacology and Experimental Therapeutics); Hans Mittelmann (Arizona State University, School of Mathematical and Statistical Sciences);
Short Abstract: The prediction of peptide epitopes for class II MHC has important biomedical applications in vaccine discovery and in understanding autoimmune diseases. Computational methods are needed since the large number of MHC variants make comprehensive experimental characterization of their peptide binding specificities infeasible. We have developed both sequence-based and structure-based prediction methods with complementary advantages. The sequence-based method, called RTA, uses a physical binding model that accounts for multiple peptide binding conformations. The RTA model also incorporates a regularization penalty that reduces overfitting. A comparison on common data sets showed that RTA outperformed several competing prediction methods. The structure-based method uses peptide docking to an MHC protein followed by machine learning based scoring using binding energy components as input to predict peptide binding affinities. Unlike sequence-based methods, it is applicable to multiple diverse MHC types. This is demonstrated by achieving significant discrimination of epitopes for four MHC types with different peptide binding preferences than the one used to fit the prediction model.
Long Abstract:Click Here

Poster W35
Structural classification of the catalytic domain of DNA polymerases using a standard measure to assess structural alignments from different methods.
Alex Slater- Pontificia Universidad Católica de Chile
Javier Castellanos (Pontificia Universidad Católica de Chile, Departamento de Genética Molecular y Microbiología. Laboratorio de Bioinformatica Molecular); Francisco Melo Ledermann (Pontificia Universidad Católica de Chile, Departamento de Genética Molecular y Microbiología. Laboratorio de Bioinformatica Molecular);
Short Abstract: DNA polymerases are a highly diverse family in terms of their sequences, but with remarkable structural similarities. In this work we show a structure-based classification of 429 catalytic domains of DNA polymerases by using a new method based on dynamic programming for selecting the best superimposition out of several solutions obtained with different structural alignment.
Long Abstract:Click Here

Poster W36
Determining RNA folding pathways with minimum energy barrier
Chris Thachuk- University of British Columbia
Anne Condon (University of British Columbia, Computer Science);
Short Abstract: We consider the problem of determining the lowest energy barrier pathway between two pseudoknot-free RNA secondary structures. We present an algorithm that determines optimal direct pathways, where direct implies that a base pair from the initial structure or final structure can be removed or added, respectively, at most once.
Long Abstract:Click Here

Poster W37
Identify the metal-binding sites in proteins by the protein contact number model
Jau-Ji Lin- Bioinformatics Program, Taiwan International Graduate Program, Academia Sinica
No additional authors
Short Abstract: We have developed a profile-based method to identify the metal-binding sites in protein. By constructing a distance-dependent profile around specific atoms (N, O, S) in the residue side chain, we can accurately and efficiently measure the tendency for a residue to be a metal binding residue.
Long Abstract:Click Here

Poster W38
Clustering large domain sequence families for function transfer and structural genomics
Robert Rentzsch- UCL
David Lee (UCL, Research Department of Structural and Molecular Biology); Christine Orengo (UCL, Research Department of Structural and Molecular Biology);
Short Abstract: To ease Structural Genomics target selection and transfer protein function annotations with high confidence we have developed a novel sequence clustering protocol (GeMMA). Unlike other methods, GeMMA does not require an initial multiple sequence alignment and can be easily parameterized to use physical and virtual (cloud) HPC infrastructure. The method can therefore sub-cluster large domain sequence families with tens of thousands of members (as found in Gene3D and Pfam), while maintaining and sometimes exceeding the performance of established protocols. Apart from benchmarking results for function transfer and SGI target selection, the presentation will also discuss our findings on the relationship of sequence and function conservation within more than 2,000 homologous domain superfamilies, different heuristics introduced to reduce the high resource demands of iterative profile-profile comparison, and our recent effort to run GeMMA in a virtual compute cluster using the Amazon EC2 cloud computing platform.
Long Abstract:Click Here

Poster W39
Distribution and Extension of Conformational Diversity on proteins domains
Ezequiel Juritz- Universidad Nacional de Quilmes
Nicolás Palopoli (Universidad Nacional de Quilmes, Structural Bioinformatic Group); Sebastián Fernández Alberti (Universidad Nacional de Quilmes, Structural Bioinformatic Group); Gustavo Parisi (Universidad Nacional de Quilmes, Structural Bioinformatic Group);
Short Abstract: Knowledge of protein conformational diversity is a key feature for comprehension and adjustment of its biological functions. Here we present a protein conformational diversity database, complemented with structural, functional and physico-chemical data from other databases to study the origin and extension of the conformational diversity.
Long Abstract:Click Here

Poster W40
SCORER 2.0: Improving the prediction of coiled-coil oligomeric state
Thomas Vincent- University of Bristol / Bristol Centre for Complexity Sciences
Craig Armstrong (University of Bristol, School of Chemistry); Dek Woolfson (University of Bristol, School of Chemistry / Dpt. of Biochemistry); Peter Green (University of Bristol, Dpt. of Mathematics);
Short Abstract: We present SCORER 2.0, a rigorous mathematical implementation of the original SCORER algorithm to predict coiled-coil oligomeric state from protein sequence information alone. The SCORER 2.0 algorithm achieves a marked improvement in our ability to differentiate dimeric and trimeric coiled-coil sequences. Future implementation of SCORER 2.0 will include multi-state prediction.
Long Abstract:Click Here

Poster W41
RactIP: fast and accurate prediction of RNA-RNA interaction using integer programming
Kengo Sato- University of Tokyo
Kengo Sato (University of Tokyo, Department of Computaional Biology); Yuki Kato (Kyoto University, Bioinformatics Center, Institute for Chemical Research); Michiaki Hamada (Mizuho Information & Research Institute, Inc, Department of Science Solution); Yoshihide Watanabe (Doshisha University, Department of Mathematical Sciences); Kiyoshi Asai (University of Tokyo, Department of Computational Biology); Tatsuya Akutsu (Kyoto University, Bioinformatics Center, Institute for Chemical Research);
Short Abstract: We present a novel method called RactIP, RNA-RNA interACTion prediction using Integer Programming. We demonstrate that RactIP is at least comparably accurate, but incomparably fast compared with competitive methods for predicting joint secondary structures.
Long Abstract:Click Here

Poster W42
Insights into the DNA Damage Response: a tale of dry and wet science.
Ana Rojas- Institute for Predictive and Personalized Medicine of Cancer
Ildefonso Cases (Institute for Predictive and Personalized Medicine of Cancer, Computational Cell Biology Group); Juan Luis Rodriguez-Barbancho (Spanish National Cancer Research Center, Genomic Instability Group); Oscar Fernandez-Capetillo (Spanish National Cancer Research Center, Genomic Instability Group);
Short Abstract: Cell integrity is a fundamental biological problem. Evolution has developed a mechanism to protect the DNA that is the DNA Damage Response (DDR), a phosphorylation-based transduction cascade. Here we show that combining large-scale phosphoproteomics data, sequence analyses and comparative genomics experimentally validated has provided insights into specific DDR components.
Long Abstract:Click Here

Poster W43
Automatic Recognition of Antifreeze Protein Based On n-peptide Composition
Chin-Sheng Yu- Feng Chia University
Chih-Hao Lu (China Medical University, Graduate Institute of Molecular Systems Biomedicine);
Short Abstract: It is reported that the AFPs cannot be distinguished in easy way. The visible characters in sequence still lack due to the poor homologs in current database. Our approach provides excellent results for discrimating the AFP from others as using the structural feature did.
Long Abstract:Click Here

Poster W44
Comparison of the efficiency of different protein structure refinement techniques.
Arne Elofsson- Stockholm University
Per Larsson (Stockholm University, DBB); Björn Wallner (Stockholm University, DBB); Erik Lindahl (Stockholm University, DBB); Patrik Björkholm (Stockholm University, DBB);
Short Abstract: Several different protocols have been proposed for the
refinement of protein structures, and several methods have
been tested. All simulations were limited to 72
h, the time frame allowed for automated predictions in CASP. Of all
protocols tested, Rosetta performs best, improving roughly 30% of all models.
Long Abstract:Click Here

Poster W45
Consensus-based protein structure prediction by means of distance restraints
Marcin Skwark- Stockholm University
No additional authors
Short Abstract: The work presents a novel approach to consensus prediction of protein 3D structures. The method proposed leverages the abundance of structural data present in the compound models to provide models of high quality, improving in many cases on the contemporary methods.
Long Abstract:Click Here

Poster W46
Low-homology protein modeling
Jian Peng- Toyota Technological Institute at Chicago
Jinbo Xu (Toyota Technological Institute at Chicago, Computer Science);
Short Abstract: We present a profile-entropy dependent scoring function for low-homology protein threading. This method models correlation among various protein features and determine their relative importance according to the amount of homologous information available. Our method greatly outperforms the best profile-based method HHpred and all the top CASP8 servers on low-homology proteins.
Long Abstract:Click Here

Poster W47
Integrative Workflow Pipeline for Functional and Structural Analysis of C-type Lectins in Immunoinformatics - Elucidating the Ligand Binding Residues of CLEC5A and Its Potential Interaction with Glyca
Dong-Yup Lee- National University of Singapore/Bioprocessing Technology Institute
Geoffrey Koh (Bioprocessing Technology Institute, Bioinformatics); Say Kong Ng (Bioprocessing Technology Institute, Animal Cells); Victor Vai Tak Wong (Bioprocessing Technology Institute, Microbial Cells);
Short Abstract: Presented herein is an integrative workflow for functional and structural analysis of C-type lectins. We applied the workflow to CLEC5A, a dengue virus (DV) binding lectin, as such identifying 13 putative residues that will be experimentally mutated to validate their roles in DV interaction.
Long Abstract:Click Here

Poster W48
Efficient Fragment-free Approach to Protein Folding
Feng Zhao- Toyota Technological Institute at Chicago
Jian Peng (Toyota Technological Institute at Chicago, Computer Science); Jinbo Xu (Toyota Technological Institute at Chicago, Computer Science);
Short Abstract: We present a new probabilistic method for fragment-free protein folding. Our method can model well the local sequence-structure relationship and sample conformations in a continuous space. Tested on the 12 CASP8 free-modeling targets, our method can generate the best models for three of them and very good models for others.
Long Abstract:Click Here

Poster W49
Domain Integrity Verification of Alternative Splicing
Hedi Hegyi- Inst of Enzymology
Peter Tompa (Inst of Enzymology, Disordered ); Lajos Kalmar (Inst of Enzymology, Disordered); Tamas Horvath (Inst of Enzymology, Disordered);
Short Abstract: While all human proteins undergo alternative splicing there are only 14 proteins with 2 isoforms in PDB. The maximum insertion accommodated by an isoform of a fully ordered protein domain was 5 amino acids. We developed a method that predicts the stability and viability of splice variants with truncated domains.
Long Abstract:Click Here

Poster W50
A new probabilistic model of RNA conformational space
Jinbo Xu- Toyota Technological Institute at Chicago
Zhiyong Wang (Toyota Technological Institute at Chicago, Computer Science);
Short Abstract: We present a new probabilistic graphical model Conditional Random Fields for RNA conformation sampling. This method can model RNA structure in a continuous (instead of discrete) space and capture well sequence-structure relationship and thus sample continuous conformations using sequence information. Results indicate this method outperforms others in sampling native-like decoys.
Long Abstract:Click Here

Poster W51
Predicting the protein-binding sites in RNA using the interaction propensity of nucleotide triplets
Yanga Byun- Inha University
Sungwook Choi (Inha University, Computer Science and Engineering); Kyungsook Han (Inha University, Computer Science and Engineering);
Short Abstract: Predicting protein-binding nucleotides is a much harder problem than predicting RNA-binding amino acids. This work presents the development of a support vector machine (SVM) that uses the interaction propensity of three consecutive nucleotides (termed nucleotide triplet) and experimental results of the SVM model with an extensive dataset of protein-RNA complexes.
Long Abstract:Click Here

Poster W52
Structure Refinement of Integral Membrane Proteins Using Energy Minimization
Christopher Summa- University of New Orleans
Kapil Pothakanoori (University of New Orleans, Computer Science);
Short Abstract: In this work we demonstrate the utility of in vacuo energy minimization for the structural refinement of integral membrane proteins. Several molecular mechanics potentials are compared vis-a-vis their ability to improve near-native protein structures.
Long Abstract:Click Here

Poster W53
POLYVIEW: Web-based Platform for Protein Structure Analysis and Visualization
Alexey Porollo- University of Cincinnati
Jaroslaw Meller (University of Cincinnati, Environmental Health);
Short Abstract: We present a suite of web servers that facilitate analysis of proteins and their complexes in terms of sequence and structure-based profiles (POLYVIEW-2D), structural and functional annotation integrated with high quality 3D rendering (POLYVIEW-3D), and animation of molecular motion and ensembles of alternative conformers from experiment and computer simulations (POLYVIEW-MM).
Long Abstract:Click Here

Poster W54
Alexandra Schnoes- University of California San Francisco
Florian Lauck (University of California San Francisco, Bioengineering and Therapeutic Sciences); Shoshana Brown (University of California San Francisco, Bioengineering and Therapeutic Sciences); Patricia Babbitt (University of California San Francisco, Bioengineering and Therapeutic Sciences);
Short Abstract: We have recently shown that the accuracy of the annotations derived from automated prediction is of significant concern. These results have motivated our development of a new, user-friendly resource for misannotation identification, prediction and curation: the Enzyme Misannotation Resource (EMR).
Long Abstract:Click Here

Poster W55
Protein Annotation Enrichment using Gene Ontology Hierarchical Structure and Protein-Protein Interactions
Seshan Ananthasubramanian- University of Pittsburgh
Yanli Wang (University of Pittsburgh, Department of Biomedical Informatics); Madhavi Ganapathiraju (University of Pittsburgh, Department of Biomedical Informatics);
Short Abstract: We predict Gene Ontology (GO) terms of proteins using GO terms of its interacting partners, hierarchical structure of GO and Bayesian inference by modeling the implicit constraints placed on pairs of GO terms of interacting proteins. We extend this to prediction of sub-cellular localization proteins
Long Abstract:Click Here

Poster W56
Structure prediction of a protein channel based on probabilistic formal grammars
Witold Dyrka- Wroclaw Univeristy of Technology
Jean-Christophe Nebel (Kingston University, Faculty of Computing, Information Systems and Mathematics); Malgorzata Kotulska (Wroclaw University of Technology, Institute of Biomedical Engineering and Instrumentation);
Short Abstract: Majority of helix-helix interactions in transmembrane proteins can be accurately represented by a set of several templates, which differ in 3D shapes. We show that these configurations of helix-helix interface can be linked to sequence level patterns represented by Stochastic Context Free Grammars induced automatically by an evolutionary algorithm.
Long Abstract:Click Here

Poster W57
Structure based design of novel small molecule inhibitors to apical membrane antigen-1 of Plasmodium falciparum
Sandhya Kortagere- Drexel University College of Medicine
Sandhya Kortagere (Drexel University College of Medicine, Microbiology); Simon Cocklin (Drexel University College of Medicine, Biochemistry); Kamal laroiya (Drexel University College of Medicine, Microbiology); Joanne Morrisey (Drexel University College of Medicine, Microbiology); Lawrence Bergman (Drexel University College of Medicine, Microbiology); Akhil Vaidya (Drexel University College of Medicine, Microbiology); James Burns (Drexel University College of Medicine, Microbiology);
Short Abstract: Malaria is endemic in mostly third world countries and nearly 500 million cases of malaria are diagnosed worldwide. There is a well recognized need to design a new generation of antimalarial drugs that can combat drug resistant forms of the malarial parasite, Plasmodium falciparum. The P. falciparum genome encodes approximately 5300 proteins, and only a few have been characterized and identified as drug and vaccine targets. Apical membrane antigen-1 (AMA-1) is an essential micronemal protein that is required for invasion of host cells by the parasite but is highly polymorphic hampering the efforts to use it as a vaccine candidate. Our design strategy was to use the crystal structure of the neutralizing antibody IgNAR in complex with AMA-1 to design small molecule inhibitors using the hybrid structure based (HSB) method. The biological activity of thirteen best ranking compounds identified by the HSB method were then tested in the parasite growth inhibition assays against various strains of P. falciparum. Five compounds (A198, A561, A784, A24142 and A24143) displayed an effect at 3?M. A24142 had an EC50 of ~700nM against all four strains of P. falciparum while A561 and A24143 had EC50 ~1?M. The specificity of A24142 and A561 to P.falciparum was demonstrated using surface plasmon resonance studies. These compounds will be pursued as a lead compound in future efforts to optimize the affinity and pharmacokinetics properties. To our knowledge, this study is the first successful application of structure-based design principles to the identification of small molecule inhibitors to AMA-1.
Long Abstract:Click Here

Poster W58
ProBiS: A web server for detection of structurally similar protein binding sites
Janez Konc- National Institute of Chemistry
Janez Konc (National Institute of Chemistry, Laboratory for Molecular Modeling);
Short Abstract: A web server, ProBiS is presented which detects structurally similar sites on protein surfaces by local surface structure alignment. It compares the query protein to members of a database of protein 3D structures and detects with sub-residue precision, structurally similar sites as patterns of physicochemical properties on the protein surface. Using an efficient maximum clique algorithm, the server identifies proteins which share local structural similarities with the query protein and generates structure-based alignments of these proteins with the query. Structural similarity scores are calculated for the query protein's surface residues, and are expressed as different colors on the query protein surface. The method has been used successfully for the detection of protein-protein, protein-small ligand, and protein-DNA binding sites. The web server is freely available at
Long Abstract:Click Here

Poster W59
A Gibbs sampler for automated classification, subgroup profiling and structural-functional annotation of an entire protein class
Andrew Neuwald- University of Maryland
Andrew F. (University of Maryland, Biochemistry);
Short Abstract: For speed and sensitivity it will become increasingly important to search a query sequence against a database of protein domain profiles rather than against millions (and soon billions) of individual sequences. However, manual curation and structural-functional annotation of domain profiles is labor intensive and is dependent on the subjective judgments of the curators. To address this issue, I describe a Gibbs sampler for constructing (hierarchically, automatically and in a statistically well-defined manner) domain profiles for every significant subgroup within a large protein class. Because the sampler classifies sequences based on the functionally critical residues characteristic of each subgroup, it can identify and annotate these residues at the same time and, using other routines, link these to corresponding crystal structure features. The sampler takes as input a multiple sequence alignment of the protein class and returns a profile of the functionally divergent subgroups within that class. When classification is uncertain, the sampler can provide predictive probabilities for alternative subgroup assignments and, in a similar manner, make probabilistic assignments to functional categories based on co-classification with functionally verified proteins. Applying the sampler to P-loop GTPases illustrates how residues that are associated with divergent functions often correspond to specific (and previously unrecognized) structural components—including, for example, a 'charge-dipole pocket' characteristic of signaling pathway GTPase on/off switches and a 'glycine brace' characteristic of Rab, Rho and Ran GTPases. The sampler uses such empirically-based distinguishing features to optimally-define and annotate profiles for each functionally-divergent subgroup.
Long Abstract:Click Here

Poster W60
Structure-based redesign of selective G-protein inactivation by RGS proteins
Mickey Kosloff- Duke University Medical Center
Mickey Kosloff (Duke University Medical Center, Duke Eye Center);
Short Abstract: As a model system to study interaction specificity between large protein families, we studied the interactions of heterotrimeric G-proteins with Regulators of G-protein Signaling (RGSs). RGS proteins inactivate G-protein mediated cellular signals by allosterically accelerating their intrinsic GTPase activity. We measured the rates at which ten RGSs inactivate the G-protein Go. We observed a range of RGS activities that surprisingly did not correlate with their subfamily classification and contrasted with the prevalent assumption that RGSs interact promiscuously with members of the Gi subfamily. We then combined our experimental results with comparative structural analysis of RGS-G-protein crystal structures, electrostatic calculations using the Finite-Difference Poisson-Boltzmann method, and in silico mutagenesis. Using a consensus approach, we markedly reduced false positives and negatives and identified RGS residues that contribute significantly to interactions with G-proteins. This approach allowed us to identify which RGS residues are necessary for their function and which residues encode selectivity. We validated our analysis against previous mutagenesis studies, showing an essentially perfect correlation. We then extended this validation by mutating additional RGS residues to show that these mutants also have impaired activity. Finally and most importantly, we used our insights into RGS-G-protein interactions to redesign gain-of-function mutants for two RGSs with poor activity. These redesigned mutants had identical activity to the high-activity RGSs that were the template for this redesign. Our multidisciplinary approach provides a quantitatively framework for understanding G-protein-RGS interactions and enables structure-based redesign of protein-protein interaction specificity at the family level.
Long Abstract:Click Here

Poster W61
A Genetic Algorithm Based Model for Predicting Functional Classification of Proteins
Yasin Bakış- Abant Izzet Baysal University
No additional authors
Short Abstract: Classification of protein sequences according to their biological function is an important task in bioinformatics. Although manually curated databases are preferred over automated processing, it is expensive and time consuming process. Thus, automated method is more desirable to annotate new sequences. In this study, we introduce a Genetic Algorithm based model for prediction of functional classification of proteins. Given the multiple sequence alignment of the protein sequences, classification of proteins was compared with functional classification. The proposed method has given highly accurate results in grouping proteins into clusters as in functional classification.
Long Abstract:Click Here

Poster W62
Application and Development of Structural Bioinformatics Methods for Rational Vaccine Design
Ivelin Georgiev- National Institutes of Health
Jiang Zhu (National Institutes of Health, Vaccine Research Center); Lawrence Shapiro (Columbia University, Biochemistry); Peter Kwong (National Institutes of Health, Vaccine Research Center);
Short Abstract: Within the context of HIV-1 and other viruses, we apply structural bioinformatics methods to three areas related to vaccine design: sera analysis, structural analysis of antibodies, and immunogen design. Preliminary experimental validation confirmed the utility of structural bioinformatics in rational vaccine design efforts.
Long Abstract:Click Here

Accepted Posters

View Posters By Category
Search Posters:
Poster Number Matches
Last Name
Co-Authors Contains
Abstract Contains