Abstract:During the early stages of drug design, identifying compounds with suitable bioactivities is crucial. Given the vast array of potential drug databases, it's feasible to assay only a limited subset of candidates. The optimal method for selecting the candidates, aiming to minimize the overall number of assays, involves an active learning (AL) approach. In this work, we benchmarked a range of AL strategies with two main objectives: (1) to identify a strategy that ensures high model performance and (2) to select molecules with desired properties using minimal assays. To evaluate the different AL strategies, we employed the simulated AL workflow based on "virtual" experiments. These experiments leveraged ChEMBL datasets, which come with known biological activity values for the molecules. Furthermore, for classification tasks, we proposed the hybrid selection strategy that unified both exploration and exploitation AL strategies into a single acquisition function, defined by parameters n and c. We have also shown that popular minimal margin and maximal variance selection approaches for exploration selection correspond to minimization of the hybrid acquisition function with n=1 and 2 respectively. The balance between the exploration and exploitation strategies can be adjusted using a coefficient (c), making the optimal strategy selection straightforward. The primary strength of the hybrid selection method lies in its adaptability; it offers the flexibility to adjust the criteria for molecule selection based on the specific task by modifying the value of the contribution coefficient. Our analysis revealed that, in regression tasks, AL strategies didn't succeed at ensuring high model performance, however, they were successful in selecting molecules with desired properties using minimal number of tests. In analogous experiments in classification tasks, exploration strategy and the hybrid selection function with a constant c<1 (for n=1) and c≤0.2 (for n=2) were effective in achieving the goal of constructing a high-performance predictive model using minimal data. When searching for molecules with desired properties, exploitation, and the hybrid function with c≥1 (n=1) and c≥0.7 (n=2) demonstrated efficiency identifying molecules in fewer iterations compared to random selection method. Notably, when the hybrid function was set to an intermediate coefficient value (c=0.7), it successfully addressed both tasks simultaneously.

The present state and challenges of active learning in drug discovery

Traversing Chemical Space with Active Deep Learning: A Computational Framework for Low-data Drug Discovery

A Guide for Active Learning in Synergistic Drug Discovery

A compact review of progress and prospects of deep learning in drug discovery

Active Learning-Driven Discovery and Dynamics Simulation of Novel SQLE Inhibitors with ADMET Analysis and Molecular Modification

An Adaptive Graph Learning Method for Automated Molecular Interactions and Properties Predictions

Concepts of Artificial Intelligence for Computer-Assisted Drug Discovery

Artificial intelligence in drug discovery: recent advances and future perspectives

Deep Learning in Virtual Screening: Recent Applications and Developments

Advancing Drug Discovery with Deep Learning: Harnessing Reinforcement Learning and One-Shot Learning for Molecular Design in Low-Data Situations

Active learning approaches in molecule pKi prediction

Sample Efficient Reinforcement Learning with Active Learning for Molecular Design

Artificial intelligence facilitates drug design in the big data era

Deep Learning and Virtual Drug Screening

ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation

Human-in-the-loop active learning for goal-oriented molecule generation

MF-LAL: Drug Compound Generation Using Multi-Fidelity Latent Space Active Learning

Deep Learning Methods for Small Molecule Drug Discovery: A Survey

Optimal Molecular Design: Generative Active Learning Combining REINVENT with Absolute Binding Free Energy Simulations

Optimal Molecular Design: Generative Active Learning Combining REINVENT with Precise Binding Free Energy Ranking Simulations

On Modeling and Utilizing Chemical Compound Information with Deep Learning Technologies: A Task-oriented Approach