DIFACQUIM research group, Department of Pharmacy, School of Chemistry, Universidad Nacional Autónoma de México, Mexico City 04510, Mexico.
J Chem Inf Model. 2021 Apr 26;61(4):1550-1554. doi: 10.1021/acs.jcim.1c00045. Epub 2021 Mar 17.
The identification of protein targets of small molecules is essential for drug discovery. With the increasing amount of chemogenomic data in the public domain, multiple ligand-based models for target prediction have emerged. However, these models are generally biased by the number of known ligands for different targets, which involves an under-representation of epigenetic targets, and despite the increasing importance of epigenetic targets in drug discovery, there are no open tools for epigenetic target prediction. In this work, we introduce Epigenetic Target Profiler (ETP), a freely accessible and easy-to-use web application for the prediction of epigenetic targets of small molecules. For a query compound, ETP predicts its bioactivity profile over a panel of 55 different epigenetic targets. To that aim, ETP uses a consensus model based on two binary classification models for each target, relying on support vector machines and built on molecular fingerprints of different design. A distance-to-model parameter related to the reliability of the predictions is included to facilitate their interpretability and assist in the identification of small molecules with potential epigenetic activity. Epigenetic Target Profiler is freely available at http://www.epigenetictargetprofiler.com.
小分子蛋白靶标的鉴定对于药物发现至关重要。随着公共领域中化学基因组数据的不断增加，已经出现了多种基于配体的靶标预测模型。然而，这些模型通常受到不同靶标已知配体数量的影响，这涉及到表观遗传靶标的代表性不足，尽管表观遗传靶标在药物发现中的重要性日益增加，但目前还没有用于表观遗传靶标预测的开放工具。在这项工作中，我们引入了表观遗传靶标预测器（ETP），这是一个免费访问且易于使用的网页应用程序，可用于预测小分子的表观遗传靶标。对于查询化合物，ETP 预测其在 55 种不同表观遗传靶标上的生物活性谱。为此，ETP 使用基于每个靶标两个二进制分类模型的共识模型，该模型依赖于支持向量机，并基于不同设计的分子指纹构建。包括与预测可靠性相关的模型参数距离，以促进其可解释性，并帮助识别具有潜在表观遗传活性的小分子。表观遗传靶标预测器可在 http://www.epigenetictargetprofiler.com 免费获得。
