深度HDAC3i:利用基于深度学习的可解释框架加速发现HDAC3抑制剂
IEEE transactions on computational biology and bioinformatics
|August 29, 2025
概括
我们开发了DeepHDAC3i,一个新的深度学习框架,用于仅使用分子结构数据识别基因脱乙酶3抑制剂 (HDAC3i). 这种工具准确地预测HDAC3i,有助于癌症治疗的发展.
科学领域:
- 遗传学和计算药物发现.
背景情况:
- 表观遗传修饰可以调节基因表达而不会改变DNA序列.
- 基因脱乙酶 (HDAC) 抑制剂在癌症治疗中使用,但缺乏特异性.
- 需要选择性HDAC抑制剂来改善癌症治疗结果.
研究的目的:
- 开发一个新的,可解释的深度学习框架,DeepHDAC3i,用于精确的HDAC3抑制剂 (HDAC3i) 的体识别.
- 仅使用SMILES标记来识别抑制剂,而无需3D连接体结构.
主要方法:
- 使用五种分子编码方法 (CDKExt,KR,KRC,Pubchem,RDKit) 来从SMILES标记中提取多视图特征.
- 使用弹性网进行最佳特征选择和1D卷积神经网络 (1D-CNN) 进行模型构建.
- 杆式的沙普利添加式扩展用于特征解释性.
主要成果:
- 在独立测试组中,DeepHDAC3i获得了高性能,精度为0.965,MCC为0.930,AUC为0.985.
- 与传统机器学习和深度学习模型相比,表现出优异的性能,精度,F1,AUC和MCC显著提高.
- 该框架为驱动HDAC3i识别的关键特征提供了可解释的见解.
结论:
- DeepHDAC3i是一种高度准确和可解释的深度学习工具,用于识别HDAC3抑制剂.
- 该框架为癌症治疗中药物发现提供了成本高效和快速的方法.
- 深度HDAC3i的性能优于现有的方法,为精确的HDAC3i识别提供了有价值的工具.
相关概念视频
Enzyme Inhibition
72.4K
Inhibitors are molecules that reduce enzyme activity by binding to the enzyme. In a normally functioning cell, enzymes are regulated by a variety of inhibitors. Drugs and other toxins can also inhibit enzymes. Some inhibitors bind to the enzyme’s active site, while others inhibit enzymatic activity by binding to other sites on the protein structure.
72.4K
Dipeptidyl Peptidase 4 Inhibitors
1.1K
Dipeptidyl peptidase 4 (DPP-4) is a serine protease widely distributed in the body. It's involved in the inactivation of GLP-1 and GIP hormones, which are crucial for insulin regulation. DPP-4 inhibitors, such as sitagliptin (Januvia), saxagliptin (Onglyza), linagliptin (Tradjenta), alogliptin (Nesina), and vildagliptin (Galvus), help increase the proportion of active GLP-1, enhancing insulin secretion. These inhibitors work by competitively binding to DPP-4. This binding causes a...
1.1K


