机器学习与生物信息学

后基因时代，探索和解释隐藏在分子生物学数据库中的有用信息将是对生物信息学研究人员的巨大挑战!为了解决分子生物学中遇到的这些难题，有效及廉价的方法是非常必要的．机器学习是一种自动的、具有智能学习技术的方法，有助于达到上述目的。本文就知识发现、人工神经网络、决策树、贝叶斯网络、遗传算法、隐马尔科夫链、聚类、归纳逻辑编程、支持向量机等机器学习方法在生物信息学中的应用进行了系统地评述．这些方法有助于加速生物分子结构预测、基因发现、基因组学和蛋白组学等方面的研究进展．

MACHINE LEARNING AND BIOINFORMATICS

:Exploring and explaining the knowledge hidden in the biomolecular database has become the grand challenge for bioinformatics in the post genome era. An efficient and inexpensive approach is required to solve problems in molecular biology; machine learning which is an automatic and intelligent learning technique may help to a-chieve this role. KDD,ANNs, Decision Trees, BBNs, GAs, HMMs, Clustering, ILP, SVM are introduced in the context of their application in bioinformatics, to experimental biologists and bioinformaticians in this paper. These approaches help to accelerate several major researches (biomolecular structure prediction, gene finding, genomics and proteomics).