Polaris Logo
DatasetsBenchmarksCompetitionsModelsBlog
Sign In

The Antiviral Competition ResultsSee the results of the Polaris competition, organized by ASAP Discovery and OpenADMET.

Polaris Logo

Built by Polaris

Privacy PolicyTerms of Use
All competitions

Hosted by

ASAP Discovery Logo

Open ADMET

Antiviral Drug Discovery 2025

ASAP Discovery is approaching a patent disclosure for its preclinical candidates for its two coronavirus Mpro drug discovery programs. This competition will mirror some of the real-world drug discovery challenges that ASAP has had to overcome in the last three years.

Ended
View ResultsHost your ownShare your Feedback

Explore the challenges

Ended 2 months ago
antiviral-ligand-poses-2025
Since the rise of structure-informed drug discovery in the 1980s-1990s, structural biology is key to drug discovery. We challenge you to predict MERS-CoV Mpro and SARS-CoV-2 Mpro poses using knowledge from the SARS-CoV-2 Mpro crystallography data that ASAP created.
A
asap-discovery
Ended 2 months ago
antiviral-potency-2025
Likely the most popular prediction for computational modellers. Although it's an important feature, as we can see in ASAP's TCP it only tells a part of the story. You will predict potency for both Mpro targets.
A
asap-discovery
Ended 2 months ago
antiviral-admet-2025
Absorption-Distribution-Metabolism-Excretion-Toxicology (ADMET) endpoints sit in the middle of the assay cascade and can make or break preclinical candidate molecules. For this blind challenge we selected several crucial endpoints for the community to predict.
A
asap-discovery

The Results

Comparing Methods

We compared methods using bootstrapping and the Tukey HSD test, presenting results via Compact Letter Display (CLD). While acknowledging that bootstrapping likely underestimates variance, we are not aware of better sampling techniques that fit the challenge format. A more robust protocol will be employed in the forthcoming publication about this challenge.
Learn more

Ligand Pose

Since the rise of structure-informed drug discovery in the 1980s-1990s, structural biology is key to drug discovery. We challenge you to predict MERS-CoV Mpro and SARS-CoV-2 Mpro poses using knowledge from the SARS-CoV-2 Mpro crystallography data that ASAP created.

NameContributors
CLD
Success Rate (<2Å)
Average RMSD
References

averkova-nika98

a
86.378 ± 2.466
1.330 ± 0.266
Code Report

ernestglukhov

b
85.363 ± 2.603
1.368 ± 0.265
Code Report

vemikainen

b
85.213 ± 2.601
1.388 ± 0.267
Code Report

vemikainenalt

c
84.673 ± 2.670
1.432 ± 0.266
Code Report

wiwnopgm

d
83.657 ± 2.769
1.549 ± 0.269
Code Report
Baseline

Potency

Likely the most popular prediction for computational modellers. Although it's an important feature, as we can see in ASAP's TCP it only tells a part of the story. You will predict potency for both Mpro targets.

Task
NameContributors
CLD
Mean Absolute Error
Mean Squared Error
Pearson R
Spearman Rho
R²
Kendall τ
References

oscar-mendez-lucio

a
0.509 ± 0.022
0.514 ± 0.064
0.789 ± 0.028
0.816 ± 0.018
0.594 ± 0.048
0.626 ± 0.016
Code Report

ccorbi

b
0.516 ± 0.022
0.529 ± 0.068
0.813 ± 0.029
0.823 ± 0.020
0.603 ± 0.051
0.649 ± 0.018
Code Report

siat353

b
0.517 ± 0.021
0.515 ± 0.062
0.788 ± 0.029
0.792 ± 0.021
0.598 ± 0.044
0.614 ± 0.018
Code Report

xubf

bc
0.518 ± 0.022
0.537 ± 0.067
0.794 ± 0.027
0.798 ± 0.020
0.594 ± 0.046
0.612 ± 0.018
Code Report

alx-dga

c
0.522 ± 0.022
0.536 ± 0.062
0.800 ± 0.027
0.811 ± 0.019
0.588 ± 0.042
0.628 ± 0.017
Code Report
Baseline

ADMET

Absorption-Distribution-Metabolism-Excretion-Toxicology (ADMET) endpoints sit in the middle of the assay cascade and can make or break preclinical candidate molecules. For this blind challenge we selected several crucial endpoints for the community to predict.

Task
NameContributors
CLD
Mean Absolute Error
Mean Squared Error
Pearson R
Spearman Rho
R²
Kendall τ
References

prairiewarbler10

a
0.224 ± 0.009
0.104 ± 0.009
0.802 ± 0.030
0.722 ± 0.024
0.623 ± 0.053
0.561 ± 0.020
Code Report

ebl88

b
0.243 ± 0.009
0.117 ± 0.009
0.783 ± 0.029
0.737 ± 0.022
0.599 ± 0.039
0.570 ± 0.018
Code Report

vchupakhin

c
0.269 ± 0.010
0.140 ± 0.010
0.731 ± 0.033
0.670 ± 0.026
0.523 ± 0.048
0.512 ± 0.021
Code Report

longhung25

d
0.276 ± 0.010
0.142 ± 0.011
0.783 ± 0.025
0.688 ± 0.024
0.546 ± 0.049
0.517 ± 0.020
Code Report

xiaolinpan

de
0.277 ± 0.010
0.144 ± 0.010
0.723 ± 0.035
0.640 ± 0.027
0.486 ± 0.047
0.494 ± 0.022
Code Report
Baseline

Host your competition on Polaris

If you're looking to release new data to the community and are interested in doing so through a competition, reach out! We'd love to collaborate.

Get in touch
A
E
V
V
W
A
J
X
O
A
S
L
X
P
C
E
V