De Novo Molecular Design Enabled by Direct Preference Optimization and Curriculum Learning

• URL: http://arxiv.org/abs/2504.01389v1
• Date: Wed, 02 Apr 2025 06:00:21 GMT
• Title: De Novo Molecular Design Enabled by Direct Preference Optimization and Curriculum Learning
• Authors: Junyu Hou,
• Abstract summary: De novo molecular design has extensive applications in drug discovery and materials science.<n>The vast chemical space renders direct molecular searches computationally prohibitive, while traditional experimental screening is both time- and labor-intensive.<n>Direct Preference Optimization (DPO) from NLP uses molecular score-based sample pairs to maximize the likelihood difference between high- and low-quality molecules.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: De novo molecular design has extensive applications in drug discovery and materials science. The vast chemical space renders direct molecular searches computationally prohibitive, while traditional experimental screening is both time- and labor-intensive. Efficient molecular generation and screening methods are therefore essential for accelerating drug discovery and reducing costs. Although reinforcement learning (RL) has been applied to optimize molecular properties via reward mechanisms, its practical utility is limited by issues in training efficiency, convergence, and stability. To address these challenges, we adopt Direct Preference Optimization (DPO) from NLP, which uses molecular score-based sample pairs to maximize the likelihood difference between high- and low-quality molecules, effectively guiding the model toward better compounds. Moreover, integrating curriculum learning further boosts training efficiency and accelerates convergence. A systematic evaluation of the proposed method on the GuacaMol Benchmark yielded excellent scores. For instance, the method achieved a score of 0.883 on the Perindopril MPO task, representing a 6\% improvement over competing models. And subsequent target protein binding experiments confirmed its practical efficacy. These results demonstrate the strong potential of DPO for molecular design tasks and highlight its effectiveness as a robust and efficient solution for data-driven drug discovery.

Related papers

• Collaborative Expert LLMs Guided Multi-Objective Molecular Optimization [51.104444856052204]
  We present MultiMol, a collaborative large language model (LLM) system designed to guide multi-objective molecular optimization.<n>In evaluations across six multi-objective optimization tasks, MultiMol significantly outperforms existing methods, achieving a 82.30% success rate.
  arXiv  Detail & Related papers  (2025-03-05T13:47:55Z)
• Diversity-Aware Reinforcement Learning for de novo Drug Design [2.356290293311623]
  Fine-tuning a pre-trained generative model has demonstrated good performance in generating promising drug molecules. No study has examined how different adaptive update mechanisms for the reward function influence the diversity of generated molecules. Our experiments reveal that combining structure- and prediction-based methods generally yields better results in terms of molecular diversity.
  arXiv  Detail & Related papers  (2024-10-14T12:25:23Z)
• Fragment-Masked Molecular Optimization [37.20936761888007]
  We propose a fragment-masked molecular optimization method based on phenotypic drug discovery (PDD) PDD-based molecular optimization can reduce potential safety risks while optimizing phenotypic activity, thereby increasing the likelihood of clinical success. The overall experiments demonstrate that the in-silico optimization success rate reaches 94.4%, with an average efficacy increase of 5.3%.
  arXiv  Detail & Related papers  (2024-08-17T06:00:58Z)
• Aligning Target-Aware Molecule Diffusion Models with Exact Energy Optimization [147.7899503829411]
  AliDiff is a novel framework to align pretrained target diffusion models with preferred functional properties. It can generate molecules with state-of-the-art binding energies with up to -7.07 Avg. Vina Score.
  arXiv  Detail & Related papers  (2024-07-01T06:10:29Z)
• Regressor-free Molecule Generation to Support Drug Response Prediction [83.25894107956735]
  Conditional generation based on the target IC50 score can obtain a more effective sampling space. Regressor-free guidance combines a diffusion model's score estimation with a regression controller model's gradient based on number labels.
  arXiv  Detail & Related papers  (2024-05-23T13:22:17Z)
• Improving Targeted Molecule Generation through Language Model Fine-Tuning Via Reinforcement Learning [0.0]
  We introduce an innovative de-novo drug design strategy, which harnesses the capabilities of language models to devise targeted drugs for specific proteins. Our method integrates a composite reward function, combining considerations of drug-target interaction and molecular validity.
  arXiv  Detail & Related papers  (2024-05-10T22:19:12Z)
• Mol-AIR: Molecular Reinforcement Learning with Adaptive Intrinsic Rewards for Goal-directed Molecular Generation [0.0]
  Mol-AIR is a reinforcement learning-based framework using adaptive intrinsic rewards for goal-directed molecular generation. In benchmark tests, Mol-AIR demonstrates superior performance over existing approaches in generating molecules with desired properties.
  arXiv  Detail & Related papers  (2024-03-29T10:44:51Z)
• CELLS: Cost-Effective Evolution in Latent Space for Goal-Directed Molecular Generation [23.618366377098614]
  We propose a cost-effective evolution strategy in latent space, which optimize the molecular latent representation vectors. We adopt a pre-trained molecular generative model to map the latent and observation spaces. We conduct extensive experiments on multiple optimization tasks comparing the proposed framework to several advanced techniques.
  arXiv  Detail & Related papers  (2021-11-30T11:02:18Z)
• Optimizing Molecules using Efficient Queries from Property Evaluations [66.66290256377376]
  We propose QMO, a generic query-based molecule optimization framework. QMO improves the desired properties of an input molecule based on efficient queries. We show that QMO outperforms existing methods in the benchmark tasks of optimizing small organic molecules.
  arXiv  Detail & Related papers  (2020-11-03T18:51:18Z)
• Guiding Deep Molecular Optimization with Genetic Exploration [79.50698140997726]
  We propose genetic expert-guided learning (GEGL), a framework for training a deep neural network (DNN) to generate highly-rewarding molecules. Extensive experiments show that GEGL significantly improves over state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-04T05:01:26Z)
• Deep Learning for Virtual Screening: Five Reasons to Use ROC Cost Functions [80.12620331438052]
  deep learning has become an important tool for rapid screening of billions of molecules in silico for potential hits containing desired chemical features. Despite its importance, substantial challenges persist in training these models, such as severe class imbalance, high decision thresholds, and lack of ground truth labels in some datasets. We argue in favor of directly optimizing the receiver operating characteristic (ROC) in such cases, due to its robustness to class imbalance.
  arXiv  Detail & Related papers  (2020-06-25T08:46:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.