paper-with-me

홈 › Papers

Lo-Hi: Practical ML Drug Discovery Benchmark

2023-10-10 · NeurIPS 2023 11 · Simon Steshin

Finding new drugs is getting harder and harder. One of the hopes of drug discovery is to use machine learning models to predict molecular properties. That is why models for molecular property prediction are being developed and tested on benchmarks such as MoleculeNet. However, existing benchmarks are unrealistic and are too different from applying the models in practice. We have created a new practical \emph{Lo-Hi} benchmark consisting of two tasks: Lead Optimization (Lo) and Hit Identification (Hi), corresponding to the real drug discovery process. For the Hi task, we designed a novel molecular splitting algorithm that solves the Balanced Vertex Minimum $k$-Cut problem. We tested state-of-the-art and classic ML models, revealing which works better under practical settings. We analyzed modern benchmarks and showed that they are unrealistic and overoptimistic. Review: https://openreview.net/forum?id=H2Yb28qGLV Lo-Hi benchmark: https://github.com/SteshinSS/lohi_neurips2023 Lo-Hi splitter library: https://github.com/SteshinSS/lohi_splitter

📄 PDF Abstract BibTeX arXiv:2310.06399

Code (2)

steshinss/lohi_neurips2023 pytorch
steshinss/lohi_splitter

Tasks

Drug DiscoveryMolecular Property PredictionProperty Prediction

Similar Papers 제목 키워드 기반

CandidateDrug4Cancer: An Open Molecular Graph Learning Benchmark on Drug Discovery for Cancer

2022-03-02 · Xianbin Ye, Ziliang Li, Fei Ma, Zongbi Yi 외

Anti-cancer drug discoveries have been serendipitous, we sought to present the Open Molecular Graph Learning Benchmark, named CandidateDrug4Cancer, a challenging and realistic benchmark dataset to facilitate scalable, ro…

Drug DiscoveryGraph Learning

MMAI Gym for Science: Training Liquid Foundation Models for Drug Discovery

2026-03-03 · Maksim Kuznetsov, Zulfat Miftahutdinov, Rim Shayakhmetov, Mikolaj Mizera 외 arxiv

General-purpose large language models (LLMs) that rely on in-context learning do not reliably deliver the scientific understanding and performance required for drug discovery tasks. Simply increasing model size or introd…

Activity PredictionDrug Discovery

DrugPlayGround: Benchmarking Large Language Models and Embeddings for Drug Discovery

2026-02-11 · Tianyu Liu, Sihan Jiang, Fan Zhang, Kunyang Sun 외 arxiv

Large language models (LLMs) are in the ascendancy for research in drug discovery, offering unprecedented opportunities to reshape drug research by accelerating hypothesis generation, optimizing candidate prioritization,…

Drug Discovery

DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaboration

2024-11-24 · Sizhe Liu, Yizhou Lu, Siyu Chen, Xiyang Hu 외

Recent advancements in Large Language Models (LLMs) have opened new avenues for accelerating drug discovery processes. Despite their potential, several critical challenges remain unsolved, particularly in translating the…

Drug Discovery

TorchDrug: A Powerful and Flexible Machine Learning Platform for Drug Discovery

2022-02-16 · Zhaocheng Zhu, Chence Shi, Zuobai Zhang, Shengchao Liu 외

Machine learning has huge potential to revolutionize the field of drug discovery and is attracting increasing attention in recent years. However, lacking domain knowledge (e.g., which tasks to work on), standard benchmar…

BIG-bench Machine LearningDrug DiscoveryMolecular Property PredictionProperty Prediction