组蛋白去乙酰化酶（HDAC）是重要的表观遗传药物靶点，在治疗癌症方面引起了科学界的广泛关注。目前市售的HDAC抑制剂缺乏对各种HDAC同工酶的选择性。在这里，我们通过药效团建模、虚拟筛选、对接、分子动力学（MD）模拟和毒理学研究等步骤，描述了我们用于发现新型潜在的羟基脂肪酸基HDAC3抑制剂的协议。建立了十种药效团假设，并通过不同的ROC（接收操作符曲线）分析验证了它们的可靠性。其中最佳的模型（假设9或RRRA）被用来搜索SCHEMBL、ZINC和MolPort数据库，以筛选出作为选择性HDAC3抑制剂的中间体，并进行不同的对接阶段。进行MD模拟（50ns）和MMGBSA研究来研究配体结合模式的稳定性，并通过轨迹分析，计算配体-受体复合物的RMSD（均方根偏差）、RMSF（均方根波动）和H键距离等。最后，对筛选出的前几种分子进行了体外毒性研究，并与参考药物SAHA和已建立的结构-活性关系（SAR）进行比较。结果表明，化合物31具有高抑制效力和较少的毒性（概率值为0.418），适合进一步的实验分析。

Histone deacetylases (HDACs) are critical epigenetic drug targets that have gained significant attention in the scientific community for the treatment of cancer. The currently marketed HDAC inhibitors lack selectivity for the various HDAC isoenzymes. Here, we describe our protocol for the discovery of novel potential hydroxamic acid based HDAC3 inhibitors through pharmacophore modeling, virtual screening, docking, molecular dynamics (MD) simulation and toxicity studies. The ten pharmacophore hypotheses were established, and their reliability was validated by different ROC (receiving operator curve) analysis. Among them, the best model (Hypothesis 9 or RRRA) was employed for searching SCHEMBL, ZINC and MolPort database to screen out hit molecules as selective HDAC3 inhibitors, followed by different docking stages. MD simulation (50 ns) and MMGBSA study were performed to study the stability of ligand binding modes and with the help of trajectory analysis, to calculate the ligand-receptor complex RMSD (root-mean-square deviation), RMSF (root-mean-square fluctuation) and H-bond distance, etc. Finally, in-silico toxicity studies were performed on top screened molecules and compared with reference drug SAHA and established structure-activity relationship (SAR). The results indicated that compound 31, with high inhibitory potency and less toxicity (probability value 0.418), is suitable for further experimental analysis.Communicated by Ramaswamy H. Sarma.