基于Transformer的蛋白质相互作用预测研究  

Transformer-based protein interaction prediction research

作　　者：靳晓宏 韦文山[1] Jin Xiaohong;Wei Wenshan(College of Electronic Information,Guangxi Minzu University,Nanning 530000,China)

机构地区：[1]广西民族大学电子信息学院,南宁530000

出　　处：《现代计算机》2024年第6期65-68,共4页Modern Computer

摘　　要：研究蛋白质相互作用(PPIs)在生物学中具有非常重要的意义,蛋白质相互作用及其相互作用类型的研究对于理解正常和疾病状态下的细胞生物学过程至关重要,从而有助于治疗靶点的识别和新药物的设计。针对传统蛋白质相互作用预测模型预测精度不够高的问题,提出了一种基于Transformer的模型TPPI对蛋白质进行相互作用的预测。五倍交叉验证实验结果表明,该模型在人类蛋白质数据集上预测的准确率为0.944,表现出了较好的预测准确性。The study of protein interactions(PPIs)is of great significance in biology,and the study of protein interactions and their interaction types is essential for understanding cell biological processes in normal and disease states,thereby aiding the iden-tification of therapeutic targets and the design of new drugs.In this paper,aiming at the problem that the prediction accuracy of tra-ditional protein interaction prediction models is not high enough,a Transformer-based model TPPI is proposed to predict protein in-teractions.The results of the five-fold cross-validation experiment show that the prediction accuracy of the model on the human pro-tein dataset is 0.944,showing good prediction accuracy.

关 键 词：TRANSFORMER 蛋白质 相互作用 预测 

分 类 号：Q811.4[生物学—生物工程]