肽和蛋白类药物靶标的生物信息学系统分析和预测

Peptide and protein drugs, including hormones, blood factors, interferons and monoclonal antibodies etc., after decades of development, currently cover a wide range of therapeutic areas and occupy an important position in pharmaceutical industry. The identification and validation of drug targets is usually the first step and also a crucial step during drug discovery, and the bioinformatics prediction of the issue that whether a protein can be used as a drug target (i.e. the prediction of protein druggability) will greatly speed up this process. However, owing to the fact that small-molecule drugs have held a dominant position in pharmaceutical industry, the bioinformatics analyses and prediction specially for the targets of peptide/protein drugs is lacking. This project will firstly systematically analyze known targets of peptide/protein drugs, and compare them with those of small-molecule drugs from multiple aspects including physicochemical properties, sequence, structure, function, network topology etc., to systematically uncover the features of peptide/protein drugs’ targets. Further based on these features, a prediction model of protein druggability for peptide/protein drugs will be developed. Finally, a genome-scale target protein list for peptide/protein drugs with detailed annotation information (including known and predicted targets) will be provided. This project may help us systematically understand the mechanism of peptide/protein drugs, speed up target identification during peptide/protein drug discovery and expand the druggable space of human genome.

肽和蛋白药物，包括激素、血液因子、干扰素和单克隆抗体等，经过几十年的发展，目前已经覆盖广泛的治疗领域，在药品市场中占据举足轻重的地位。确定并验证药物靶标往往是药物研发的第一步也是至关重要的一步，预测一个蛋白是否可以被用作药物靶标（即蛋白可药性预测）将极大地加速此过程。然而由于小分子药物一直占主导地位，至今尚无专门针对肽和蛋白药物靶标的生物信息学分析和预测工作。本课题首先从理化性质、序列、结构、功能、网络拓扑等方面对已知的肽和蛋白药物的靶标蛋白展开系统分析并同小分子药物靶标进行比较，以从整体上多方面系统揭示肽和蛋白药物靶标的特点；进而基于这些特点，发展一个针对肽和蛋白药物的蛋白可药性预测体系；最后给出一个基因组范围的注释详细的肽和蛋白药物靶标数据集（包括已知的和预测的靶标）。本课题有助于从整体上系统理解肽和蛋白药物的作用机制，加速肽和蛋白药物研发中的靶标确定环节，扩展人类基因组上的可药空间。

肽和蛋白药物，包括激素、血液因子、干扰素和单克隆抗体等，经过几十年的发展，目前已经覆盖广泛的治疗领域，在药品市场中占据举足轻重的地位。确定并验证药物靶标往往是药物研发的第一步也是至关重要的一步，预测一个蛋白是否可以被用作药物靶标将极大地加速此过程。然而由于小分子药物一直占主导地位，至今尚无专门针对肽和蛋白药物靶标的生物信息学分析和预测工作。该工作首先系统梳理现在已成功的肽、蛋白、小分子药物靶标，统计分析结果暗示和传统的小分子药物靶标相比，蛋白药物具有更强的靶标革新能力，且倾向直接靶向疾病原因蛋白；进而首次从多方面系统揭示这两类药物靶标在序列、结构、功能等方面特点，以及同传统的小分子药物靶标的不同之处；而后成功构建首个肽/蛋白药物靶标评估模型；最后提供了基因组范围的肽/蛋白药物靶标预测数据集。该工作将促进从整体上理解不同类型药物靶标的作用机制，提高药物研发中肽和蛋白药物靶标选择的效率和成功率，将有益于学术界和工业界。