Bioactivity Data

Bioactivity data analysis focuses on predicting the biological activity of molecules, primarily to accelerate drug discovery and repurposing. Current research emphasizes developing and benchmarking machine learning models, particularly deep learning architectures like graph neural networks (GNNs) and reinforcement learning algorithms, to improve the accuracy and generalizability of bioactivity predictions across diverse datasets and assay types. This work addresses challenges like data heterogeneity, covariate shift, and limited labeled data, aiming to create more robust and reliable predictive models for various applications in drug development and beyond. The ultimate goal is to reduce the time and cost associated with experimental drug screening, leading to more efficient and effective therapeutic development.

Papers

July 16, 2024

A Hitchhiker's Guide to Deep Chemical Language Processing for Bioactivity Prediction
Rıza Özçelik, Francesca Grisoni
Predictive Model Molecular Representation Practical Guide Molecular String Representation Bioactivity Prediction Bioactivity Data Chemical Language

January 30, 2024

A large dataset curation and benchmark for drug target interaction
Alex Golts, Vadim Ratner, Yoel Shoshan, Moshe Raboh, Sagi Polaczek, Michal Ozery-Flato, Daniel Shats, Liam Hazan, Sivan Ravid, Efrat Hexter
New Benchmark Drug Discovery Data Curation Drug Target Interaction Bioactivity Data

December 3, 2023

Distributed Reinforcement Learning for Molecular Design: Antioxidant case
Huanyi Qin, Denis Akhiyarov, Sophie Loehle, Kenneth Chiu, Mauricio Araya-Polo
Case Study Deep Q Network Molecular Design Novel Molecule Molecular Discovery Molecular Energy Bioactivity Data

November 27, 2023

Target-Free Compound Activity Prediction via Few-Shot Learning
Peter Eckmann, Jake Anderson, Michael K. Gilson, Rose Yu
LeArning Abstract Drug Discovery Bioactivity Prediction Bioactivity Data

August 17, 2023

Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions
Lucian Chan, Marcel Verdonk, Carl Poelking
Drug Discovery Neural Process Virtual Screening Unobserved Heterogeneity Affinity Prediction Computer Aided Drug Discovery Bioactivity Prediction Bioactivity Data

July 14, 2023

Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions
Leo Klarner, Tim G. J. Rudner, Michael Reutlinger, Torsten Schindler, Garrett M. Morris, Charlotte Deane, Yee Whye Teh
Deep Learning Drug Discovery Prior Knowledge Simple Function Covariate Shift Scoring Molecule Bioactivity Data Domain Prior Drug Discovery Task

August 3, 2022

Hybrid Approach to Identify Druglikeness Leading Compounds against COVID-19 3CL Protease
Imra Aqeel, Abdul Majid
Covid 19 Hybrid Approach Sars Cov 2 Drug Like Non Structural Protein Chemical Compound Bioactivity Data

November 27, 2021

Benchmarking Accuracy and Generalizability of Four Graph Neural Networks Using Large In Vitro ADME Datasets from Different Chemical Spaces
Fabio Broccatelli, Richard Trager, Michael Reutlinger, George Karypis, Mufei Li
Graph Neural Network Graph Convolutional Network Stronger Generalizability Benchmark Study Chemical Space Graph Neural Network Model Bioactivity Data

Bioactivity Data

Papers

A Hitchhiker's Guide to Deep Chemical Language Processing for Bioactivity Prediction

A large dataset curation and benchmark for drug target interaction

Distributed Reinforcement Learning for Molecular Design: Antioxidant case

Target-Free Compound Activity Prediction via Few-Shot Learning

Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions

Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions

Hybrid Approach to Identify Druglikeness Leading Compounds against COVID-19 3CL Protease

Benchmarking Accuracy and Generalizability of Four Graph Neural Networks Using Large In Vitro ADME Datasets from Different Chemical Spaces