Saved in:
Bibliographic Details
Main Authors: Qu, Yunni, Vaduri, Bhargav, Jatoth, Karthikeya, Wellnitz, James, Dinh, Dzung, Veenbaas, Seth, Chapman, Jonathan, Tropsha, Alexander, Oliva, Junier
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.01825
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917359836463104
author Qu, Yunni
Vaduri, Bhargav
Jatoth, Karthikeya
Wellnitz, James
Dinh, Dzung
Veenbaas, Seth
Chapman, Jonathan
Tropsha, Alexander
Oliva, Junier
author_facet Qu, Yunni
Vaduri, Bhargav
Jatoth, Karthikeya
Wellnitz, James
Dinh, Dzung
Veenbaas, Seth
Chapman, Jonathan
Tropsha, Alexander
Oliva, Junier
contents Machine learning (ML) models are increasingly deployed for virtual screening in drug discovery, where the goal is to identify novel, chemically diverse scaffolds while minimizing experimental costs. This creates a fundamental challenge: the most valuable discoveries lie in out-of-distribution (OOD) regions beyond the training data, yet ML models often degrade under distribution shift. Standard novelty-rejection strategies ensure reliability within the training domain but limit discovery by rejecting precisely the novel scaffolds most worth finding. Moreover, experimental budgets permit testing only a small fraction of nominated candidates, demanding models that produce reliable confidence estimates. We introduce EXPLOR (Extrapolatory Pseudo-Label Matching for OOD Uncertainty-Based Rejection), a framework that addresses both challenges through extrapolatory pseudo-labeling on latent-space augmentations, requiring only a single labeled training set and no access to unlabeled test compounds, mirroring the realistic conditions of prospective screening campaigns. Through a multi-headed architecture with a novel per-head matching loss, EXPLOR learns to extrapolate to OOD chemical space while producing reliable confidence estimates, with particularly strong performance in high-confidence regions, which is critical for virtual screening where only top-ranked candidates advance to experimental validation. We demonstrate state-of-the-art performance across chemical and tabular benchmarks using different molecular embeddings.
format Preprint
id arxiv_https___arxiv_org_abs_2406_01825
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Reliable OOD Virtual Screening with Extrapolatory Pseudo-Label Matching
Qu, Yunni
Vaduri, Bhargav
Jatoth, Karthikeya
Wellnitz, James
Dinh, Dzung
Veenbaas, Seth
Chapman, Jonathan
Tropsha, Alexander
Oliva, Junier
Machine Learning
Artificial Intelligence
Machine learning (ML) models are increasingly deployed for virtual screening in drug discovery, where the goal is to identify novel, chemically diverse scaffolds while minimizing experimental costs. This creates a fundamental challenge: the most valuable discoveries lie in out-of-distribution (OOD) regions beyond the training data, yet ML models often degrade under distribution shift. Standard novelty-rejection strategies ensure reliability within the training domain but limit discovery by rejecting precisely the novel scaffolds most worth finding. Moreover, experimental budgets permit testing only a small fraction of nominated candidates, demanding models that produce reliable confidence estimates. We introduce EXPLOR (Extrapolatory Pseudo-Label Matching for OOD Uncertainty-Based Rejection), a framework that addresses both challenges through extrapolatory pseudo-labeling on latent-space augmentations, requiring only a single labeled training set and no access to unlabeled test compounds, mirroring the realistic conditions of prospective screening campaigns. Through a multi-headed architecture with a novel per-head matching loss, EXPLOR learns to extrapolate to OOD chemical space while producing reliable confidence estimates, with particularly strong performance in high-confidence regions, which is critical for virtual screening where only top-ranked candidates advance to experimental validation. We demonstrate state-of-the-art performance across chemical and tabular benchmarks using different molecular embeddings.
title Reliable OOD Virtual Screening with Extrapolatory Pseudo-Label Matching
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2406.01825