Are you sure you want to leave this community? Leaving the community will revoke any permissions you have been granted in this community.
Slides and videos for presentations from the March 27 D3R Webinar are now available here.
We are pleased to invite you to join the D3R Webinar, 8am - 2pm Pacific Time, Monday, March 27. The agenda is here; topics include: reports on the PL-2016-1 and GC2 challenges refinement of co-crystal structures plans for future Grand Challenges and CELPP Please pre-register via this link. You will then be sent a confirmation email. Please follow the "check system requirements" link in that email, to make sure you have the required plugin and are ready to go before the meeting starts.
More updates, and an invitation!
Sharing of GC2 ProtocolsWe have a number of updates and questions for you this week! Sharing of GC2 Protocols Some participants have asked for information on other participants' computational protocols, and we plan to share these, coded by submission IDs, on the D3R website. If you are participating anonymously and are concerned that your protocol will de-anonymize you, please let us know as soon as possible. Special Issue of JCAMD for your GC2 paper You are invited to submit a paper about your GC2 calculations to another special issue of the Journal of Computer-Aided Molecular Design. (Thanks to editor Terry Stouch!). We anticipate a submission deadline of June 1, 2017. Instructions for submitting your article are provided here. Citable DOIs for Datasets Digital Object Identifiers (DOIs) have been created for the Grand Challenge 2015 datasets and are available on the Datasets page. We are working on DOIs for the GC2 datasets and hope to have them available in time to be cited in your in your JCAMD paper. Grand Challenge 2 Webinar, March 27 We plan to invite a number of participants to present their experiences with GC2 and PL-2016-1 at the webinar, with time for Q/A. There will also be talks by Stephen Burley about Xray structures of protein-ligand cocrystal structures, and by the experimentalists who generated the challenge data. The preliminary agenda is posted here.
We have completed initial processing of all Stage 1 and Stage 2
predictions, and have posted them on the D3R website. To access them,
please go to this page, log in, and click the link under Evaluation Results at the top right.
This will take you to a page with downloadable CSV files and web-based
graphical analyses that put your predictions into the context of others.
You can find your own results by its submission ID.
The page also summarizes how the statistics were computed and how
compounds measured as racemates were dealt with in the evaluations. The
information on racemates is also provided in this document.
We are happy to report that there were 124 submissions from 33
participants for Stage 2 of Grand Challenge 2! The answer file of
measured IC50 values is available here (direct download here).
Pat Walters has generously agreed to lead the evaluation process again.
Evaluation results will be made available as they become available.
We will also be sending more information about the March 27 Webinar to
discuss this challenge and the PL-2016-1 mini-challenge. We hope to
interact with you there!
Several participants have asked how to make acceptable four-character PDB and MOL file names from the 5-character PDB IDs provided at the close of Stage 1.
To do this, just leave the "1" off the beginning of the name. For example, 1HGAF.pdb would become HGAF.pdb, leading to file names like HGAF-FXR_1.mol and HGAF-FXR_1.pdb.
1. In response to a participant request, and in light of the late
release of corrected Stage 1 cocrystal structures, the deadline for
Stage 2 predictions is extended by one week, to February 8.
2. Summary evaluations of your Stage 1 pose predictions are available here. Look for your results by Receipt ID.As
you could submit up to five poses for each ligand, we list Mean RMSD of
Top Scoring Pose (the mean, across all ligands, of the poses you listed
as the top scoring ones) and Mean RMSD of Lowest RMSD Poses (the mean,
across all ligands, of the lowest RMSD poses of the five for each
ligand). The RMSDs were computed by aligning your predictions against
the new, re-refined crystal structures, using binding-site atoms to do
the superpositions. Ligand-by-ligand statistics will appear on the D3R
website soon.
In case you would like to investigate the Stage 1 crystal structures in more detail, we have posted the mtz files here. (Note that this is a 99MB download.)