CA-DEL: An Open Multi-Target, Multi-Modal Benchmark for Learning from DNA-Encoded Library Screens

TL;DR

CA-DEL is a new public benchmark for machine learning in drug discovery, focusing on predicting selectivity among homologous enzymes using noisy sequencing data and validated binding affinities.

Contribution

It introduces a multi-modal, multi-target benchmark with real binding affinity data, addressing the challenge of noisy DEL screening signals in drug discovery models.

Findings

Benchmark includes data for three homologous carbonic anhydrase isoforms.

Integrates sequencing read counts with high-fidelity binding affinity data.

Facilitates development of models that generalize from noisy to accurate biophysical measurements.

Abstract

The success of machine learning in drug discovery hinges on learning the relationship between a chemical structure and its biological activity. While DNA-Encoded Library (DEL) technology can generate the massive datasets required for this task, its primary signal -- sequencing read counts -- is an indirect and often noisy proxy for true molecular binding affinity. To address the scarcity of public benchmarks for developing robust models that can overcome this data challenge, we introduce CA-DEL, a multi-dimensional public benchmark featuring screens against three homologous carbonic anhydrase isoforms. While recent benchmarks like KinDEL have introduced 3D poses for kinase targets, CA-DEL distinguishes itself by focusing on the selectivity challenge among homologous Carbonic Anhydrase isoforms (CAII, CAIX, CAXII). Unlike benchmarks relying solely on noisy enrichment scores, CA-DEL integrates a rigorous validation set of experimentally determined binding affinities ($K_i$) from ChEMBL, establishing a critical Sim-to-Real evaluation paradigm: training on noisy DEL screens and testing on high-fidelity biophysical data.