DONATE
To be eligible for consideration this Application form must be submitted on cache-challenge.org by 23:59 CET on 25th February 2024.
Please note that by submitting this Application and, if accepted for participation, by making one or more Submissions to this CACHE Challenge, you are agreeing on behalf of yourself and your organization to the CACHE Terms of Participation, found here: CACHE_Terms_of_Participation.
Please fill in all fields in this form in order to be eligible to participate.
APPLICATIONS REVIEW: Applications will be pre-screened at CACHE to ensure that the Applicant research group or company has published in the field or otherwise has appropriate qualifications. Applications will then go through a double-blind peer-review process where the Computational Approach Description(s) submitted below will be available to the reviewers. Each Applicant that passes the pre-screen stage will be asked to review the quality of the Computational Approach from five other Applicants (without identifying Applicants) and to verify that the Computational Approach Descriptions are sufficiently detailed that an expert in the field could readily understand the Applicants’ prediction strategies. Applications that pass the double-blind peer-review stage will be reviewed by an independent committee convened by CACHE that will select up to 25 Applicants to participate and make submissions of binding predictions to this CACHE Challenge.1 All selected Participants will be anonymized.
1. CACHE has no obligation to accept any particular Application but, resources permitting, will strive to accept all Applications with viable Computational Approach Descriptions and appropriate Applicant qualifications, as determined by CACHE in its discretion.
THE CHALLENGE: The Challenge will be composed of three rounds:
[1] hit identification: Participants will select or design up to 100 chemically novel antagonists with an IC50 for MCHR1 below 30 µM. While close analogs to known ligands will not be dismissed, chemical novelty will be one of the metrics used by the independent Hit Evaluation Committee.
[2] virtual screening of merged selections: Participants will receive a file with the chemical structures of all hit candidates selected by all Participants in step [1], and will predict up to 100 active compounds.
[3] hit optimization: Participants who have at least one hit-candidate from step [1] experimentally validated with an IC50 for MCHR1 below 30 µM will select or design up to 50 follow-up compounds
RELEASING THE DATA: All Computational Approach Descriptions associated with a particular step of the challenge ([1] hit identification; [2] virtual screening of merged selections; [3] hit optimization) will be publicly released immediately after all predictions for this particular step are submitted. For steps [1] and [3] of the challenge, all experimental data on all compounds from all Participants, but without the structure of the compounds, will be publicly released immediately after the data was generated for all Participants. Participants will be able to link the data to compound structures, but only for their compounds. Molecular weight and cLogP ranges of compounds will be publicly released. Immediately after all Participants have submitted their prediction for step [2] of the challenge and the activity of all compounds has been determined experimentally, the status (“active” or not) predicted by all Participants will be compared with the experimental activity and publicly released. Compound structures will remain undisclosed. All data and all compound structures will be publicly released at the end of the challenge, at least three months after the experimental data are shared with the relevant Participant.
At the end of the challenge, the three top-performing Participants will be de-anonymized and their identities publicly released, as well as all Participants that request to be publicly identified, and Participants that withdrew before the end of the challenge.
To preserve the integrity of the challenge, Participants must keep confidential their compound structures and the performance of their specific Computational Approaches (including the Participants’ identity in relation to specific experimental data) until the end of the challenge, after CACHE has publicly released all data and compound structures. CACHE will provide instructions to Participants regarding permitted statements by participants during the course of the challenge (e.g. Participants may be permitted to state publicly the number of hits in a particular round of the challenge, etc.).
Please note that by entering this challenge, pursuant to the CACHE Terms of Participation of this CACHE Challenge (see above), you and your organization are agreeing that none of the compounds submitted to CACHE will be patented by you or your organization and are instead intended as public domain resources. Your algorithms, software code, and reproducible workflows, including any improvements you make to them through participation in this CACHE Challenge, are your Intellectual Property and no licenses or rights are granted by you in them (unless you have made them open source and indicated as such below).
Bryan Roth’s laboratory at University of North Carolina where all experimental assays will be conducted will cover 100% of the costs of testing compounds experimentally. For this Challenge, CACHE will cover 25% of the costs (up to $2500) of procuring make-on-demand compounds (but not custom-synthesized compounds) for eligible academics and companies with less than 20 employees, and 50% of the costs (up to $5000) for participants whose entire computational pipeline is open source (no proprietary software). Each participant will be required to remit payment to Enamine directly for the participant’s portion of compound procurement costs.
The information in this section will be maintained confidential unless you click otherwise below, or are in the top three performing Participant’s in this CACHE Challenge.
Details of the corresponding applicant (to whom emails will be sent):
If your application is accepted:
You will have two months starting March 20th 2024, with a Submission deadline May 20th 2024, to select or design up to 100 chemically novel antagonists with an IC50 for MCHR1 below 30 µM. Compounds with cLogP > 5 or MW > 550 Da will be filtered out. Compounds should be commercially available at modest cost. Compounds should be available from the Enamine Real Database, subsets of which may be downloaded from http://cache.docking.org/ or obtained directly from Enamine. Please provide SMILES and purchasing information for each compound (Enamine ID, etc).
Alternatively, participants can have compounds custom-synthesized at their own cost.
CACHE will submit to Enamine CACHE’s portion of the payment for your selected compounds, and you will be required to pay Enamine directly for the remainder.
Compounds will then be tested in Dr. Bryan Roth’s laboratory at University of North Carolina in a competition assay with a tritiated 19-mer cyclic peptide agonist, and potent molecules validated in this biophysical assay will be tested in a functional cellular assay to evaluate their antagonist activity. All experimental data will be sent back to you, barring unforeseen circumstances, by October 31st 2024. To save costs, procured molecules will not necessarily include stereochemical resolution. The lowest cost form of the compound available will be procured; a salt form will be requested for solubility if available, unless otherwise informed.
In addition to submitting their predictions by May 20th 2024, Participants using de novo design approaches at their own cost will be responsible for having their predicted compounds and associated analytical spectra showing >95% purity received at CACHE by August 1st 2024.
This section will be used for double-blind peer-review and will be disclosed on cache-challenge.org anonymously (except as described above) after all participants have submitted their first compound selection:
By June 20th 2024, barring unforeseen circumstances, you will receive an SDF file with the 2D structure of all compounds selected by all Participants at the hit identification phase, but without any experimental data. You will be asked to predict up to 100 actives among these compounds by August 20th 2024.
This section will be used for double-blind peer-review and will be disclosed on cache-challenge.org anonymously (except as described above) after all merged selection participants have submitted their merged selection:
By October 31st 2024, barring unforeseen circumstances, you will have received all experimental data for the compounds that you selected that were successfully delivered to CACHE. From among the hits (IC50 < 30 µM) that emerge from your work, you will be asked to select by December 31st 2024 an additional 50 follow-up compounds from Enamine to explore the structure-activity relationship of these experimental hits. Your file must contain the Enamine catalogue numbers and SMILES string of each compound as well as the parent compound (which will be used as positive control). CACHE will submit to Enamine CACHE’s portion of the payment for as many of your selected compounds as can be made economically by parallel synthesis at Enamine, and you will be required to pay Enamine directly for the remainder. CACHE will then test them experimentally, measure the solubility and purity of the highest affinity hits, and send all data back to you, including the IC50 of active molecules by July 1st 2025. If this second selection is not from the Enamine Real Database, participants will be responsible for synthesis at their own cost and for having the compounds and associated analytical spectra showing >95% purity received at CACHE by April 20th 2025.
This section will be used for double-blind peer-review and will be disclosed on cache-challenge.org anonymously (except as described above) after all hit-optimization participants have submitted their selection of optimized compounds
On October 1st 2025, barring unforeseen circumstances, CACHE will publicly release the anonymized list of participants and their origin (academia, biotech, pharma etc.). As described above, Participant identity will be de-anonymized if they are a top three performer or have consented to release of their identity above. Information about a Participant’s software and/or reproducible workflows will be released if the Participant has consented above. For all Participants the following will be released:
• Computational Method Description(s) • Structures of selected compounds and associated experimental data • Virtual screening of merged selection
Methods will be evaluated and ranked using a traffic-light scoring scheme detailed in Ackloo, S., et al. Nat Rev Chem 6, 287-296 (2022). As noted above, please note that by submitting this Application and, if accepted for participation, by making one or more Submissions to this CACHE Challenge, you are agreeing on behalf of yourself and your organization to the CACHE Terms of Participation, found here: CACHE Terms of Participation.
