Cunningham Michael, Pins Danielle, Dezső Zoltán, Torrent Maricel, Vasanthakumar Aparna, Pandey Abhishek
Genomics Research Center, AbbVie Inc., 1 North Waukegan Rd., North Chicago, IL, 60064, USA.
Information Research, AbbVie Inc., 1 North Waukegan Rd., North Chicago, IL, 60064, USA.
J Cheminform. 2023 Jul 19;15(1):64. doi: 10.1186/s13321-023-00735-7.
The identification of human proteins that are amenable to pharmacologic modulation without significant off-target effects remains an important unsolved challenge. Computational methods have been devised to identify features which distinguish between "druggable" and "undruggable" proteins, finding that protein sequence, tissue and cellular localization, biological role, and position in the protein-protein interaction network are all important discriminant factors. However, many prior efforts to automate the assessment of protein druggability suffer from low performance or poor interpretability. We developed a neural network-based machine learning model capable of generating druggability sub-scores based on each of four distinct categories, combining them to form an overall druggability score. The model achieves an excellent performance in separating drugged and undrugged proteins in the human proteome, with an area under the receiver operating characteristic (AUC) of 0.95. Our use of multiple sub-scores allows the assessment of potential protein targets of interest based on distinct contributors to druggability, leading to a more interpretable and holistic model to identify novel targets.
识别可通过药物调节且无显著脱靶效应的人类蛋白质，仍然是一个重要的未解决挑战。人们已经设计出计算方法来识别区分“可成药”和“不可成药”蛋白质的特征，发现蛋白质序列、组织和细胞定位、生物学作用以及在蛋白质-蛋白质相互作用网络中的位置都是重要的判别因素。然而，许多先前自动化评估蛋白质可成药性的努力都存在性能低下或可解释性差的问题。我们开发了一种基于神经网络的机器学习模型，该模型能够根据四个不同类别中的每一个生成可成药性子分数，并将它们组合形成一个总体可成药性分数。该模型在区分人类蛋白质组中已用药和未用药蛋白质方面表现出色，受试者操作特征曲线下面积（AUC）为0.95。我们使用多个子分数，能够基于可成药性的不同贡献因素评估潜在的感兴趣蛋白质靶点，从而形成一个更具可解释性和整体性的模型来识别新靶点。
