Gómez-Sacristán Pablo, Simeon Saw, Tran-Nguyen Viet-Khoa, Patil Sachin, Ballester Pedro J
Centre de Recherche en Cancérologie de Marseille, Marseille 13009, France.
NanoBio Laboratory, Widener University, Chester, PA 19013, USA.
J Adv Res. 2025 Jan;67:185-196. doi: 10.1016/j.jare.2024.01.024. Epub 2024 Jan 26.
Small-molecule Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD1/PDL1) inhibition via PDL1 dimerization has the potential to lead to inexpensive drugs with better cancer patient outcomes and milder side effects. However, this therapeutic approach has proven challenging, with only one PDL1 dimerizer reaching early clinical trials so far. There is hence a need for fast and accurate methods to develop alternative PDL1 dimerizers.
We aim to show that structure-based virtual screening (SBVS) based on PDL1-specific machine-learning (ML) scoring functions (SFs) is a powerful drug design tool for detecting PD1/PDL1 inhibitors via PDL1 dimerization.
By incorporating the latest MLSF advances, we generated and evaluated PDL1-specific MLSFs (classifiers and inactive-enriched regressors) on two demanding test sets.
60 PDL1-specific MLSFs (30 classifiers and 30 regressors) were generated. Our large-scale analysis provides highly predictive PDL1-specific MLSFs that benefitted from training with large volumes of docked inactives and enabling inactive-enriched regression.
PDL1-specific MLSFs strongly outperformed generic SFs of various types on this target and are released here without restrictions.
通过程序性死亡配体1（PDL1）二聚化来抑制小分子程序性细胞死亡蛋白1/程序性死亡配体1（PD1/PDL1），有望开发出价格低廉、能为癌症患者带来更好疗效且副作用更小的药物。然而，这种治疗方法已被证明具有挑战性，到目前为止，只有一种PDL1二聚体药物进入了早期临床试验。因此，需要快速且准确的方法来开发其他的PDL1二聚体药物。
我们旨在证明，基于PDL1特异性机器学习（ML）评分函数（SFs）的基于结构的虚拟筛选（SBVS）是一种强大的药物设计工具，可通过PDL1二聚化来检测PD1/PDL1抑制剂。
通过纳入最新的MLSF进展，我们在两个严格的测试集上生成并评估了PDL1特异性MLSF（分类器和富含非活性化合物的回归模型）。
生成了60个PDL1特异性MLSF（30个分类器和30个回归模型）。我们的大规模分析提供了具有高度预测性的PDL1特异性MLSF，这些模型受益于大量对接的非活性化合物的训练以及富含非活性化合物的回归分析。
在这个靶点上，PDL1特异性MLSF在性能上大大优于各种类型的通用SF，并且在此无限制发布。
