Newest papers

PMO-Dock: Benchmarking Docking, Specificity, and Generalization in Molecular Optimization

Gor Simonyan, Tatevik Abrahamyan, Narek Abelyan, Tigran Fahradyan, Hrant Khachatrian · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.19011 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

The Practical Sampling Efficiency Optimization (PMO) benchmark evaluasi standar dalam optimasi molekuler, tetapi dibangun pada oracle sederhana berbasis properti yang tidak menangkap desain obat berbasis struktur. Sebagai bidang telah pindah ke tujuan berbasis docking, praktik evaluasi telah menjadi tidak standar, dan metode yang berbeda melaporkan hasil pada tugas-tugas docking yang berbeda, sering tanpa protokol bersama untuk perbandingan yang adil atau keterbatasan transfer sampel-efisiensi. Untuk mengatasi ini, kami memperkenalkan Practical Molecular Optimization for Docking (PMO-Dock), sebuah benchmark dan protokol untuk optimasi berbasis docking yang terdiri dari 25 tugas yang mencakup hit generation, lead optimization, dan tugas spesifikasi baru yang membutuhkan ikatan target-target yang kuat sementara menghukum interaksi off-target. Tugas protokol ini memisahkan

Machine translation (Meta M2M100); the English original follows.

Abstract

The Practical Molecular Optimization (PMO) benchmark standardized evaluation in molecular optimization, but it is built on simple property-based oracles that do not capture structure-based drug design. As the field has moved to docking-based objectives, evaluation practice has become unstandardized, and different methods report results on different docking tasks, often without a shared protocol for fair comparison or sample-efficiency constraints. To address this, we introduce Practical Molecular Optimization for Docking (PMO-Dock), a benchmark and protocol for docking-based optimization consisting of 25 tasks covering hit generation, lead optimization, and a new specificity task requiring strong on-target binding while penalizing off-target interactions. The protocol separates model development from final evaluation: validation tasks are used for hyperparameter tuning, and test tasks are held out for final evaluation and are not used during development. Evaluation on the test tasks is done under a strict oracle budget to reflect realistic drug-discovery settings. We benchmark four diverse high-performing methods spanning different optimization paradigms, Saturn (reinforcement learning), GenMol (discrete diffusion), Genetic-guided GFlowNet, and ChemLlama (LLM-based evolutionary search). Our results show no universally best method across tasks, additional analyses of hyperparameter transfer (cross-task and cross-category) and sensitivity show that transfer is often limited and varies by method, motivating method-dependent selection strategies. This benchmark provides a concrete standard for evaluating generalization to unseen tasks and sample efficiency under a limited test-time oracle budget.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Benchmark (surveying), Oracle, Benchmarking, Machine learning, Generalization

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange