TDRtargets.org : an open-access resource for prioritizing possible drug targets and linking them to possible inhibitors 11001011010100010010110111101100110011010110 01001110110110001101010110010110001010010101 Gregory J. Crowther 1 and Fernán Agüero 2 with Santiago J. Carmona 2 , M. Paula Magariños 2 , Dhanasekaran Shanmugam 3 , Maria A. Doyle 4 , Christiane Hertz-Fowler 5 , Matthew Berriman 5 , Solomon Nwaka 6 , Stuart A. Ralph 4 , David S. Roos 3 , John P. Overington 7 , and Wesley C. Van Voorhis 1 1 University of Washington, 2 Universidad de San Martín, 3 University of Pennsylvania, 4 University of Melbourne, 5 Wellcome Trust Sanger Institute, 6 TDR / World Health Organization, and 7 European Bioinformatics Institute
23
Embed
TDRtargets.org: an open-access resource for prioritizing possible drug targets and linking them to possible inhibitors
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
TDRtargets.org: an open-access resource for prioritizing possible drug targets
• Original goal: facilitate identification of proteins with traits of good drug targets.
predicted from protein binding pockets and similarities to known
drug targets (A. Hopkins, B. Al-Lazikani, J. Overington)
orthology is used to make inferences about incompletely studied proteins (D. Roos)
according to sigma.com and brenda-enzymes.orgaccording to Protein
Data Bank (pdb.org)
Weighting allows proteins to be ranked based on many criteria without discarding those that lack some desired criteria; e.g.,
1. Protein Y (75 points)2. Protein Z (45 points)3. Protein X (30 points)
Overview of TDRtargets.org
• Original goal: facilitate identification of proteins with traits of good drug targets.
Sample Criterion
Weight Protein X Protein Y Protein Z
Assayable 20 Yes Yes Yes
Crystallizable 10 Yes No No
Druggable 30 No Yes No
Essential 25 No Yes Yes
A gene page
Examples of prioritizing targets
“Identification of attractive drug targets in neglected-disease pathogens using an in silico approach” (G. J. Crowther et al., PLoS Negl. Trop. Dis. 4: e804, 2010)
• made good lists of promising drug targets in several species (http://www.tdrtargets.org/published/browse/366)
• compared to lists previously published by others
• explored plusses and minuses of bioinformatics-based rankings
Figure 2: A summary of the multiparameter search queries presented in this study.
Criterion Weight
is a protein 1
has associated PubMed publications 20
has a solved crystal structure 20
has a ModBase 3D model 10
has a druggability index ≥ 0.4 20
has a compound desirability index > 0.2 10
has a precedent for assayability 20
classified by KEGG as a glycolytic/gluconeogenic enzyme 1000
glycolytic flux control (based on M. A. Albert et al., 2005) glyceraldehyde-3-phosphate dehydrogenase (1.2.1.12)
– Possible explanations (non-exclusive):• Targets found through phenotypic screens but do not meet usual criteria
for target-based approach• Assumption that loss-of-function phenotype is best• Total pool of viable targets greatly exceeds the clinically validated ones
• Ranks #2 in Table 4 of PLoS NTD paper• Nonessential for blood-stage growth!• Significance of low, not-tightly-regulated expression during blood stage?
– M. tuberculosis pantothenate kinase (PanK or CoaA)• Ranks in top 100 of Table 5 of PLoS NTD paper• Screens found potent enzyme inhibitors, but none kill cells (C. E. Barry)• Enzyme activity vastly exceeds what is required for growth (C. E. Barry)
• No list is canonical– Researchers have legitimate differences of opinion
• Helminths: penalize proteins with human orthologs, or not?• M. tuberculosis: target information-processing proteins?
– Rankings should change as new data arrive– Make your own!
Emerging challenges in drug discovery
• How can we link “cell-active” compounds (discovered in whole-cell screens) to specific targets?
• How can we study novel proteins that don’t have known inhibitors?
► Importance of compound-target links! ◄
Recent and forthcoming progress on TDRtargets.org:1. Add infrastructure for handling chemical data.
2. Expand the number of bioactive compounds in the database.
3. Link compounds with targets (via literature curation and informatics).
• Coming up: predictions based on docking simulations, compound similarities, etc.
2° associations
What would make TDRtargets.orgeven more useful and popular?
• More screening data (e.g., for M. tuberculosis)?
• Additional ways to link compounds and targets?
• Additional datasets (e.g., transcriptomics) for prioritizing targets, and better user interface via closer alignment with EuPathDB.org?
• Other ideas?
2° association . . .Upgrade to 1°?
Summary
• TDRtargets.org is an open-access database that facilitates target-based drug development for neglected diseases.
• Targets may be prioritized with weighted searches of multiple criteria.
• The main goal of the website is NOT to establish “canonical” top-10 lists, but to let visitors use their own criteria to find targets that are attractive to them.
• A focus of ongoing work is the use of curation and informatics to link compounds and targets.