Challenges

D3R aims to engage the CADD community through blind prediction challenges, training opportunities and hosting an annual workshop. The blind challenges will be based on the previously unpublished datasets donated by our collaborators, augmented where needed by D3R, and made available through this website. Through our workshops we will work to integrate efforts with other CADD community initiatives, such as SAMPL and TDT, provide training seminars on a number of dataset/CADD related topics, and develop a forum for CADD issue discussions.

Grand Challenges

Calculation of protein-ligand poses and affinities or scores. Each dataset centers on a specific protein target, and includes crystal structures and affinity data for multiple drug-like ligands. Applicable computational methods may include fast docking and scoring, energy calculations based on quantum chemistry methodologies, and simulation-based free energy methods.

SAMPL

Calculations of chemical properties and binding affinities for model systems, as well as informative protein-ligand systems not fitting the Grand Challenge schema. Model challenges include hydration free energies of small molecules, aqueous-organic partition coefficients, and host-guest binding thermodynamics.

Teach-Discover-Treat

Challenges developed around various real-life computational chemistry scenarios, primarily for targets in neglected diseases. In addition to assessing predictive ability, participants are required to submit comprehensive, step-by-step, tutorials describing their end-to-end workflows.