TDR Targets: a chemogenomics resource for neglected diseases

AGÜERO F

Florianopolis

Conferencia; 7th International Conference of the Brazilian Association for Bioinformatics and Computational Biology (X-meeting); 2011

Associação Brasileira de Bioinformática e Biologia Computacional (AB3C)

TDR Targets: a chemogenomics resource for neglected diseases María Paula Magariños1, Santiago J Carmona1, Gregory J Crowther2, Stuart A Ralph3, David S Roos4, Dhanasekaran Shanmugam4, Wesley Van Voorhis2 and Fernán Agüero1. 1 Instituto de Investigaciones Biotecnológicas, Universidad de San Martín, San Martín, Buenos Aires, Argentina. 2 Department of Medicine, University of Washington, Seattle, Washington, USA 3 Bio21 Molecular Science and Biotechnology Institute, The University of Melbourne, Victoria, Australia 4 Department of Biology and Penn Genomics Institute, University of Pennsylvania, Philadelphia, USA The TDR Targets Database (http://tdrtargets.org) is an online resource that seeks to facilitate the prioritization of molecular targets for drug development, allowing users to numerically weight the evidence available for each gene. The focus of the database is on pathogens that are in the portfolio of the World Health Organization Tropical Disease Research Programme , which currently includes Mycobacterium tuberculosis/leprae, Trypanosoma brucei/cruzi, Leishmania major, Schistosoma mansoni, and Brugia malayi and its endosymbiont Wolbachia. From the computational standpoint, the database is an exercise in data integration, allowing inferences to be made from the underlying data. Developed as a MySQL database, with an accompanying Catalyst (Perl MVC) web application, TDR Targets integrates pathogen specific genomic information with functional data (e.g., expression, phylogeny, essentiality) for genes, collected from various sources. These include an in-house literature curation effort focused on the annotion of validation data relevant for drug discovery. This annotation uses an ontology-based pheno-syntax to describe phenotypes associated with genetic and/or chemical experiments. In the current release of TDR Targets we integrated various chemical datasets, associating them with protein targets based on manual curation of the literature. Compounds were obtained from DrugBank, PubChem and Starlite (ChEMBL) databases. A cheminformatics pipeline was developed in-house to calculate physicochemical properties and identifiers for each molecule, used to facilitate searches and cross-linking to other databases. Using this pipeline, we have now integrated 825,814 compounds. Bioactive compounds obtained from the Starlite/ChEMBL database have information on their activity (IC50s, MICs, etc.) and their targets. Using the OrthoMCL database of orthologous proteins, the database allows the identification of pathogen proteins that share the same ortholog group with at least one target in ChEMBL, therefore transitively linking these targets to candidate chemical leads. These recent additions to the database greatly facilitate the exploration of linkages (both curated and inferred) between genes and small molecules, effectively allowing the navigation of chemical space in a genomics context. Keywords: TDR Targets, drug discovery, neglected diseases, chemogenomics Supported by Agencia Nacional de Promoción Científica y Tecnológica (PICT-2010-1479), Special Programme for Research and Training in Tropical Diseases (TDR, OMS A50830), and the Fogarty International Center (Grant Number D43TW007888).