facilitated by Christopher Austin (NIH), Jerome Hert (UCSF), Gerard Kleywegt (University of Uppsala), Xavier Barril (University of Barcelona), Barry Hardy (Douglas Connect), Darryl Reid (SimBioSys), Paul Hawkins (OpenEye Scientific Software), Kay Perry (University of Pennsylvania)and other practitioners

This workshop will have an agenda to:

• share experiences on current practices for comparative studies in virtual screening
• engage in group-based discussion and detailed documentation and agreement of best practices for conducting comparison studies
• to co-create requirements for a comparison study which will subsequently be conducted by the community of practice

The agenda of workshop will be designed so as to maximise interaction, discussion, issue resolution, and action plans for cooperation. Workshop activities will address the specific challenges:

• statistically significant relationships between docking scores and ligand affinity
• practices and procedures for the operation of community-based screening and docking comparisons including tests and interpretation of results, in a way that everyone can agree is fair.
• use of wiki-based approaches for practice development
• peer review, data compilation, running of programs, judgement of results
• workflow descriptions for comparisons
• beyond conformational energetics in the rank ordering of diverse compounds in high throughput virtual screening
• measurement and benchmarking
• binding mode prediction, virtual screening for lead identification, rank-ordering by affinity for lead optimization
• atom typing, ligand preparation (ionic forms, tautomers, ...), ligand conformer generation, protein preparation (protonation, residue orientation, ...), ligand placement (top-down, bottom-up, fragment based, group based, ...), energy calculation (force field type, grid type, algorithm, ...), constraint handling (global and local optimization strategy? process to escape local minima?), scoring (single-objective, multi-objective, consensus, ...)
• separation of test set information from model development
• validation datasets, results and applicability domains
• objective comparisons of standardized test datasets
• extraction of data from the scientific literature
• methods and procedures for secure testing of commercial data that could be acceptable to industry
• frameworks for computational model testing and validation
• impact of knowledge management approaches
• collaboration and community support structures and environments

Workshop Format

• Virtual communication and collaboration approaches will be used pre- and post-event to maximise the benefit of the workshop. In particular a wiki will be opened 3 months prior to the workshop to commence group documentation of supporting materials and to start to populate the workshop program with initial suggestions, ideas and methods. The wiki will only be open to practice group and workshop participants, although the group can subsequently agree to publish selected material from the created content more openly at a subsequent date, once a level of quality has been reached and the rules for specific study comparison agreed.
• During the workshop extensive use of facilitated small group discussions using a knowledge café format will be used.
• The results of all discussions will be collected by workshop facilitators and entered into the practice group wiki, to support subsequent group activities and development initiatives, including future practice group meetings and research and development projects.

Registration required:

Yes

Free:

No