BRUSELAS: A HPC based software architecture for drug discovery on large molecular databases Antonio-Jesús Banegas-Luna Horacio Pérez-Sánchez Alberto Caballero Bioinformatics and High Performance Computing Research Group (BIO-HPC) Universidad Católica San Antonio de Murcia (UCAM) 1 Thursday, November 30th 2017 Nombre de la presentación Universidad Católica San Antonio de Murcia - Tlf: (+34) 968 27 88 00 info@ucam.edu - www.ucam.edu Nombre del profesor - Tlf: (+34) 968 00 00 00 - mail@pdi.ucam.edu
1. Introduction Virtual Screening (VS) is the use of high performance computing to analyze large databases of chemical compounds in order to identify possible drug candidates. In-vivo screening Expensive in time, money and manpower. Not feasible verifying all the known compounds. In-silico screening or Virtual screening In-silico search of drug-like candidates. Performed before in-vivo screening. Saves time, money and manpower. Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
2. Why VS is a Big-Data issue? Volume • Compounds • Conformers • Interactions • Space on disk Veracity • • Statistics Compound • VS vs Experimental properties • • Validation Biological • In vivo testing activities 5Vs of • • Predictions Publications • Ligands Big Data • Targets Variety • Software • Computation time • Drug designing • Processes • New predictions • Multithreading • Extensive to other • Heuristics diseases Velocity • Populate Value databases Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
3. Proposed solution: Drawbacks & needs Difficult integration of VS tools Multiple task to handle Several combinations to explore Low flexibility in experiments Easy tooling integration Easy interpretation of results High level of customization Accuracy & Speed Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
3. Proposed solution: BRUSELAS BRUSELAS: B alanced R apid and U nrestricted S erver for E xtensive L igand- A imed S creening. http://bio-hpc.ucam.edu/Bruselas Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
4. BRUSELAS Experiment configuration Experiment configuration Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
4. BRUSELAS Result explorer Result explorer Download Smiles of query Download results in mol2/CSV formats Share results by email Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
4. BRUSELAS Result explorer Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
4. BRUSELAS Compound explorer Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
4. BRUSELAS Outlook and future works Advanced HPC techniques Application to Find new other scopes, drug e.g. farming, candidates nutraceutical or cosmetics Combination New features and with AI and compounds genetics 10 Nombre de la presentación Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu. Nombre del profesor - Tlf: (+34) 968 00 00 00 - mail@pdi.ucam.edu
4. BRUSELAS Collaborations We search for groups and companies to collaborate in publications and its application to real projects. Contact: Antonio J. Banegas ajbanegas@gmail.com Horacio Pérez Sánchez hperez@ucam.edu https://www.bsc.es/hpc-europa3-hosts Next H2020 Health Tetramax SBIR 11 Nombre de la presentación Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu. Nombre del profesor - Tlf: (+34) 968 00 00 00 - mail@pdi.ucam.edu
5. HPC projects in BIO-HPC 2017 Eurolab-4-HPC Business Prototyping Project Proposals: Call 2017. Funded by Horizon H2020-FETHPC-2014 (GA 671610). 2016 Eurolab-4-HPC Business Prototyping Project Proposals: Call 2016. Funded by Horizon H2020-FETHPC-2014 (GA 671610). 2015 TETRACOM Technology Transfer in Computing Systems (funded by the European Union Commission Framework Programme 7 - Contract No. 609491), “ Advanced Computational Drug Discovery Technologies using High Performance Computing Architectures (ACDDT HPC)” . 2013 PRACE Distributed European Computing Initiative (DECI-10), “Novel Anticoagulants ” . 500000 GPU hours supercomputing time. 2011 HPC-Europa2, “High Throughput in-silico screening in HPC architectures for the discovery of new inhibitors against blood diseases ” . 25000 hours supercomputing time. 2010 CESGA - Singular Science And Technology Infrastructure, “High Throughput in-silico screening in HPC architectures for the discovery of new inhibitors against blood diseases ” . 200000 hours supercomputing time. 2010 Distributed European Infrastructure Supercomputing Applications, “BLOODINH” . PI: Wolfgang Wenzel. 2M hours supercomputing time. 2008 Marie Curie FP7 IEF “INSILICODRUGDISCOVER”, ID:220147. Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
6. References J. Bajorath , “ Integration of virtual and high-throughput screening ”, Nat. Rev. Drug. Discov. 1(11), 882-894 (2002). H. Pérez-Sánchez, V. Rezaei, V. Mezhuyev, D. Man, J. Peña-García, H. Den-Haan, S. Gesing, “ Developing science gateways for drug discovery in a grid environment ”, Springerplus, 5(1), 1300 (2016) D. Laney, “3D Data management: controlling data volume, velocity and variety”, Appl. Deliv. Strateg, Internet (February 2001):1 – 4 (2001) W.P. Walters, M.T. Stahl and M.A. Murcko, "Virtual Screening-An Overview", Drug Discovery Today, 3, 160-178 (1998) W.H. Shin, X. Zhu, M. Bures, D. Kihara, "Three-dimensional compound comparison methods and their application in drug discovery", Molecules, 20(7), 12841-12862 (2015) DIA-DB, http://bio-hpc.ucam.edu/dia-db/index.php, Bioinformatics and High Performance Computing Research Group (2015) A.S. Reddy, S.P. Pati, P.P. Kumar, H.N. Pradeep, G.N. Sastry, "Virtual screening in drug discovery - a computational perspective", Current Protein & Peptide Science, 8(4), 329-351 (2007) Bioinformatics and High Performance Research Group (BIO-HPC) - http://bio-hpc.eu.
Recommend
More recommend
