Outlines Drug discovery HTS Screening Computational Screening - PowerPoint PPT Presentation

Modeling HTS Screening for Drug discovery Ying-Ta Wu Genomics Research Center, Academia Sinica, Taiwan. e-mail: ywu@gate.sinica.edu.tw

Outlines • Drug discovery • HTS Screening • Computational Screening • Structure-based approach • Compound-based approach • Large-Scale Virtual Screening on GRID • Bioactive Compound Profile

Drug Discovery • Disease Targets New Drug • Chemical Entity ------ > drug New Candidate New Lead New Target ~14yrs New Gene

Drug Targets Human Non-Human parasites, bacteria, viruses, … Nature review drug discovery

Drug Development by chemist intuition … slay the dragon ! sword ax dagger armor G.L. Patrick An Introduction to Medicinal Chemistry, Oxford University Press, 1995

Drug Discovery by chemist intuition … Target R2 O N R3 SAR N Assay (structure Activity relationship) R1 optimization cycle Candidate

Time and Cost How many compounds can a chemist synthesize per years? -- few hundreds R2 R2 O N NH 2 R3 H O O R4 N O + + X H 2 N R3 R1 R1 how many? 10 9 Library – chemical landscape 10 100 Screening – speed vs cost Solution: automation and miniaturize. HTS Alternative? VS

New Paradigm in Drug Discovery Development candidate is taken forward Development candidate is taken forward Development candidate is taken forward Development candidate is taken forward Target structure obtained Target structure obtained Target structure obtained Target structure obtained Chemistry begins Chemistry begins Chemistry begins Chemistry begins HTS hits confirmed HTS hits confirmed HTS hits confirmed HTS hits confirmed Screen Strategy HTS HTS HTS HTS Assay developed Assay developed Assay developed Assay developed Target selected Target selected Target selected Target selected Refer to Walters et al. DDT, 3, 160-178 (1998)

New Paradigm in Drug Discovery Homology modeling Homology modeling Homology modeling Homology modeling Database clustering Database clustering Database clustering Database clustering Similarity analysis/ Similarity analysis/ Similarity analysis/ Similarity analysis/ Virtual screening Virtual screening Virtual screening Virtual screening QSAR QSAR QSAR QSAR Pharmacophores Pharmacophores Pharmacophores Pharmacophores library library library library Structure-based design/ Structure-based design/ Structure-based design/ Structure-based design/ selecting selecting selecting selecting lead optimizing lead optimizing lead optimizing lead optimizing 2-4 years 2-4 years 2-4 years 2-4 years Development candidate is taken forward Development candidate is taken forward Development candidate is taken forward Development candidate is taken forward Target structure obtained Target structure obtained Target structure obtained Target structure obtained Chemistry begins Chemistry begins Chemistry begins Chemistry begins HTS hits confirmed HTS hits confirmed HTS hits confirmed HTS hits confirmed HTS HTS HTS HTS Assay developed Assay developed Assay developed Assay developed Target selected Target selected Target selected Target selected Refer to Walters et al. DDT, 3, 160-178 (1998)

Drug Discovery by Screening Strategy Example: Broad (random) Screen reconfirmation reconfirmation HTS HTS primary hits primary hits library library assay assay cpd 1 cpd 1 cpd 1 cpd 1 cpd 2 cpd 2 cpd 2 cpd 2 confirmed hits confirmed hits cpd 3 cpd 3 cpd 3 cpd 3 cluster/MCS/mode cluster/MCS/mode hit series hit series cluster 1 cluster 1 cluster 1 single single single single single single cluster 2 cluster 2 cluster 2 cluster 2 cluster 2 cluster 2 selected hits selected hits prioritized hits prioritized hits SAR/ADME/IP SAR/ADME/IP substructure substructure extended hits extended hits repository repository similarity similarity cluster 3 cluster 3 cluster 3

Drug Discovery by Screening Strategy Random Screen (> 1,000,000) • whole library screening at initial stage • no biased, novel • need uHTS if library is large • low hit rate, cost Focused Screen (~10,000) • specific collection screening • manageable, efficient • need prior bioactive information • novelty Sequential Screen (5000~10,000) • representative subset or explore hit series, which may be recruited after other two screen procedures • need clustering, data-mining, etc. • initial selection

Sequential / Focused Screen • SD docking virtual screening • LB filtering, similar searching medicinal chemistry • subset diversity (features) lead opt Initial library HTS data analysis • clustering new library active model HD HA θ Z

Computational modeling When to apply Information drives drug discovery -- the more of it, the sooner, the better. What methods and tools virtual library, cluster, screen, or score…

Examples : computational methods and tools • Target structure-based • Compound-based • Score and cluster

Study Case Develop inhibitor of SARS coronavirus main protease-3CL pro

A: Pre-screening NCI 25K compounds commercial available modified RO5 structural diversity 2K compounds

B: Docking Methods create energy maps/element on active-site Docking Engine: AutoDock 3.0.5 Garrett M. Morris David S. Goodsell Ruth Huey William E. Hart evaluate 1.5 x 10 6 energies/molecule Scott Halliday Rik Belew Arthur J. Olson Morris et al. (1998), J. Computational Chemistry , 19 : 1639-1662. 1. Prepare the Target Protein -- add polar hydrogen atoms carry out 50 x 10 runs/molecule -- assign charges to atoms -- decide range of binding site 2. Run AutoGrid cluster on RMSD=1Å and max  G binding 3. Prepare the Ligand -- assign charges to atoms -- decide flexible bonds (run AutoTors) 4. Run AutoDock test compounds with 5. Evaluate Results and Rank Score -  G binding > 12 kcal/mol

Structure-Based Design: Modeling Thiazin derivatives H N N H Example: R2 N N S HN O S HN O NH R1 HO 26a O NH R3 HO -13 kcal/mol 3x10^ -10 26a O -13 kcal/mol 3x10^ -10 Cys-145 Glu_166

Case Result H H H H H N N N N N N COOCH 3 N N N H N N S S S S S CN 40 uM JMF312 30 uM JMF310 10 uM 10 4 uM JMF311 CH 3 CN H H H H N N N N N N N N S Br S CH 3 S S JMF314 JMF315 JMF316 JMF313 37 uM 17 uM 21 uM 16 uM CF 3 H OCH 3 NO 2 H N N CH 3 H H N N N N N N S S S S JMF320 JMF318 JMF319 18 uM JMF317 > 50 uM > 50 uM 23 uM N N H H H H N N N N N N N N N CH 3 OCH 3 S S S S JMF322 JMF321 JMF323 JMF309 15 uM 20 uM 33 uM > 50 uM

Screening by Molecular Docking

Application Characteristics • Computational screening based on molecular docking is the most time consuming part in structure-based drug design workflow • The requirement of CPU power and storage space increases proportional to the number of compounds and target proteins involved in the screening Number of docking tasks = N x M – N: number of compounds – M: number of target structures The Challenge • CPU-bound application, huge amount of output, no communication between tasks

Grid-Enabling Virtual Screening

DCI: against Influenza A

Replication cycle of Flaviviridae EuAsiaGrid: Dengue virus Kuhn, R.J. et al . Cell 108 , 717−725; 2002 http://en.wikipedia.org/wiki/Aedes