Ali Khodabandeh Yalabadi, Mehdi Yazdani-Jahromi, Ozlem Ozmen Garibay
{"title":"BoKDiff:最佳的k扩散对准目标特定的3D分子生成。","authors":"Ali Khodabandeh Yalabadi, Mehdi Yazdani-Jahromi, Ozlem Ozmen Garibay","doi":"10.1093/bioadv/vbaf137","DOIUrl":null,"url":null,"abstract":"<p><strong>Motivation: </strong>Structure-based drug design (SBDD) leverages the 3D structure of target proteins to guide therapeutic development. While generative models like diffusion models and geometric deep learning show promise in ligand design, challenges such as limited protein-ligand data and poor alignment reduce their effectiveness. We introduce BoKDiff, a domain-adapted framework inspired by alignment strategies in large language and vision models that combines multi-objective optimization with Best-of-K alignment to enhance ligand generation.</p><p><strong>Results: </strong>Built on DecompDiff, BoKDiff generates diverse ligands and ranks them using a weighted score based on QED, SA, and docking metrics. To overcome alignment issues, we reposition each ligand's center of mass to match its docking pose, enabling more accurate sub-component extraction. We further incorporate a Best-of-N (BoN) sampling strategy to select optimal candidates without model fine-tuning. BoN achieves QED > 0.6, SA > 0.75, and over 35% success rate. BoKDiff outperforms prior models on the CrossDocked2020 dataset with an average docking score of -8.58 and 26% valid molecule generation rate. This is the first study to integrate Best-of-K alignment and BoN sampling into SBDD, demonstrating their potential for practical, high-quality ligand design.</p><p><strong>Availability and implementation: </strong>Code is available at https://github.com/khodabandeh-ali/BoKDiff.git.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbaf137"},"PeriodicalIF":2.8000,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12228967/pdf/","citationCount":"0","resultStr":"{\"title\":\"BoKDiff: best-of-K diffusion alignment for target-specific 3D molecule generation.\",\"authors\":\"Ali Khodabandeh Yalabadi, Mehdi Yazdani-Jahromi, Ozlem Ozmen Garibay\",\"doi\":\"10.1093/bioadv/vbaf137\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Motivation: </strong>Structure-based drug design (SBDD) leverages the 3D structure of target proteins to guide therapeutic development. While generative models like diffusion models and geometric deep learning show promise in ligand design, challenges such as limited protein-ligand data and poor alignment reduce their effectiveness. We introduce BoKDiff, a domain-adapted framework inspired by alignment strategies in large language and vision models that combines multi-objective optimization with Best-of-K alignment to enhance ligand generation.</p><p><strong>Results: </strong>Built on DecompDiff, BoKDiff generates diverse ligands and ranks them using a weighted score based on QED, SA, and docking metrics. To overcome alignment issues, we reposition each ligand's center of mass to match its docking pose, enabling more accurate sub-component extraction. We further incorporate a Best-of-N (BoN) sampling strategy to select optimal candidates without model fine-tuning. BoN achieves QED > 0.6, SA > 0.75, and over 35% success rate. BoKDiff outperforms prior models on the CrossDocked2020 dataset with an average docking score of -8.58 and 26% valid molecule generation rate. This is the first study to integrate Best-of-K alignment and BoN sampling into SBDD, demonstrating their potential for practical, high-quality ligand design.</p><p><strong>Availability and implementation: </strong>Code is available at https://github.com/khodabandeh-ali/BoKDiff.git.</p>\",\"PeriodicalId\":72368,\"journal\":{\"name\":\"Bioinformatics advances\",\"volume\":\"5 1\",\"pages\":\"vbaf137\"},\"PeriodicalIF\":2.8000,\"publicationDate\":\"2025-06-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12228967/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Bioinformatics advances\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1093/bioadv/vbaf137\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"MATHEMATICAL & COMPUTATIONAL BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bioinformatics advances","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/bioadv/vbaf137","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"MATHEMATICAL & COMPUTATIONAL BIOLOGY","Score":null,"Total":0}
引用次数: 0
摘要
动机:基于结构的药物设计(SBDD)利用目标蛋白的3D结构来指导治疗开发。虽然像扩散模型和几何深度学习这样的生成模型在配体设计中表现出了希望,但有限的蛋白质配体数据和较差的对齐等挑战降低了它们的有效性。我们介绍了BoKDiff,一个受大型语言和视觉模型中的对齐策略启发的领域适应框架,它结合了多目标优化和Best-of-K对齐来增强配体生成。结果:BoKDiff建立在DecompDiff的基础上,生成了不同的配体,并使用基于QED、SA和对接指标的加权分数对它们进行排名。为了克服对齐问题,我们重新定位每个配体的质心以匹配其对接姿态,从而实现更准确的子成分提取。我们进一步采用Best-of-N (BoN)采样策略来选择不需要模型微调的最优候选者。BoN达到QED > 0.6, SA > 0.75,成功率超过35%。BoKDiff在CrossDocked2020数据集上的平均对接分数为-8.58,有效分子生成率为26%,优于先前的模型。这是首个将Best-of-K对准和BoN采样整合到SBDD的研究,展示了它们在实用、高质量配体设计方面的潜力。可用性和实现:代码可从https://github.com/khodabandeh-ali/BoKDiff.git获得。
BoKDiff: best-of-K diffusion alignment for target-specific 3D molecule generation.
Motivation: Structure-based drug design (SBDD) leverages the 3D structure of target proteins to guide therapeutic development. While generative models like diffusion models and geometric deep learning show promise in ligand design, challenges such as limited protein-ligand data and poor alignment reduce their effectiveness. We introduce BoKDiff, a domain-adapted framework inspired by alignment strategies in large language and vision models that combines multi-objective optimization with Best-of-K alignment to enhance ligand generation.
Results: Built on DecompDiff, BoKDiff generates diverse ligands and ranks them using a weighted score based on QED, SA, and docking metrics. To overcome alignment issues, we reposition each ligand's center of mass to match its docking pose, enabling more accurate sub-component extraction. We further incorporate a Best-of-N (BoN) sampling strategy to select optimal candidates without model fine-tuning. BoN achieves QED > 0.6, SA > 0.75, and over 35% success rate. BoKDiff outperforms prior models on the CrossDocked2020 dataset with an average docking score of -8.58 and 26% valid molecule generation rate. This is the first study to integrate Best-of-K alignment and BoN sampling into SBDD, demonstrating their potential for practical, high-quality ligand design.
Availability and implementation: Code is available at https://github.com/khodabandeh-ali/BoKDiff.git.