基于机器学习的蛋白质相互作用文献挖掘方法研究进展

Progress of research in machine learning-based protein-protein interaction extraction

蛋白质相互作用是生物体内一类极其重要的分子活动.自动挖掘、整合生物文献中的蛋白质相互作用有助于生物学的研究,获得了人们的广泛关注,成为生物文献挖掘领域的重要任务之一.目前,基于机器学习的蛋白质相互作用挖掘方法已经取得了很大进步,对该领域的进展进行归纳总结将有助于方法的进一步优化和应用.本文在对机器学习方法构建流程介绍的基础上,进一步从机器学习的分类器、学习特征、方法评估以及挖掘系统4个方面对蛋白质相互作用文献挖掘进行系统总结,并探讨了其发展前景.

Protein-Protein Interaction（PPI） is a kind of molecular event that plays a very important role in life activities. Automatic extraction and integration of PPIs from biological literature, which is one of the most important tasks of biological literature mining, can contribute to the research of molecular biology and has received widespread attention. Currently, machine learning-based PPI extraction methods have made great progress. It will help us to promote the improvement and application of these methods by reviewing their progresses. In this review, we firstly introduce the general construction process of machine learning-based PPI extraction method. Then, we summarize the classifiers, the features and the evaluation of machine learning-based PPI extraction method and PPI extraction tools. Finally, we discuss the prospects in this field.