微调的深度转移学习模型用于对更安全的药物进行大规模查,目标是A类GPCRs
Davide Provasi1, Marta Filizola1
1Department of Pharmacological Sciences, Icahn School of Medicine at Mount Sinai, New York, New York 10029, United States.
Biochemistry
|March 8, 2025
概括
预测更安全的G蛋白结合受体 (GPCR) 药物是一个挑战. 这项研究开发了使用转移学习的AI模型来识别低效或偏差的激动剂,帮助药物发现以提高安全性.
科学领域:
- 药理学 药理学是指药理学的学科.
- 计算生物学 计算生物学
- 药物发现 药物发现 药物发现
背景情况:
- G蛋白结合受体 (GPCRs) 对于细胞信号传递至关重要,是主要的药物标.
- 将药物的疗效与特定的信号通路联系起来,并预测治疗窗口仍然很困难.
- 了解低内在疗效和带偏差是开发更安全药物的关键.
研究的目的:
- 开发预测性AI模型,用于识别具有低内在疗效或偏向激进主义的GPCR配体.
- 为了克服GPCR药物发现的深度学习中的数据限制.
- 通过预测其生物活性概况,促进发现更安全的候选药物.
主要方法:
- 在A类GPCR序列和配体数据上预训练了一个深度学习模型.
- 利用转移学习和具有自然语言处理的神经网络.
- 嵌入式受体突变对模型改进的信号传输的影响.
主要成果:
- 开发了两种微调模型:一种用于低效率激动剂,一种用于偏向激动剂.
- 单个A类GPCR的模型可按需提供.
- 启用了大型化学图书馆的虚拟选,以寻找潜在的候选药物.
结论:
- 开发的AI模型可以预测GPCR配体,从而提高安全性.
- 这些模型通过促进更安全的化合物的识别,大大推动了药物发现.
- 该方法解决了在GPCR药物开发中有限的高质量数据的挑战.
相关概念视频
Transducer Mechanism: G Protein–Coupled Receptors
1.8K
G Protein–Coupled Receptors (GPCRs) are membrane-bound receptors that transiently associate with heterotrimeric G proteins and induce an appropriate response to various stimuli. GPCRs regulate critical physiological pathways and are excellent drug targets for treating diseases such as diabetes, cancer, obesity, depression, or Alzheimer's. Nearly 35% of approved drugs implement their therapeutic effects by selectively interacting with specific GPCRs.
GPCRs are also called heptahelical,...
GPCRs are also called heptahelical,...
1.8K
Drug Discovery: Overview
7.3K
Drug discovery is a multifaceted process involving extensive screening, testing, and optimization of lead compounds to identify potential new drugs for therapeutic use. It combines several approaches, including screening large numbers of natural products, chemical modification of known active molecules, identification of new drug targets, and rational design based on biological mechanisms and drug-receptor structure. These approaches are carried out in both academic research laboratories and...
7.3K


