《生命科学》 2024, 36(3): 302-315

机器学习方法在植物基因组信息预测中的研究现状

张兆旭¹ , 管思彤¹ , 林一鸣¹ , 苏培森² , 黄思罗³ , 孟宪勇¹ , 柳平增^1,* , 颜　君^1,*

¹山东农业大学信息科学与工程学院，农业农村部黄淮海智慧农业技术重点实验室，泰安 271018 ²聊城大学农学与农业工程学院，聊城 252000 ³华中科技大学生命科学与技术学院，武汉 430074

摘　要：

随着高通量测序技术的飞速发展，植物基因组学研究目前已经积累了海量多组学数据。因此如何开发和改进相关处理软件工具，从而有效利用这些海量数据发掘有用的生物学信息，成为当下亟需解决的重要科学问题。其中机器学习方法凭借其显著的预测、分类、数据挖掘和集成能力，在此领域受到广泛关注。本文系统综述了不同类型机器学习方法的基本原理和流程，以及这些方法在植物基因组功能预测中的研究进展，重点总结了机器学习模型在植物分子相互作用预测、重要功能位点预测、功能注释、作物育种等方面的应用成果，并展望了该领域未来的发展方向和应用前景。本文有助于植物研究者快速了解和应用机器学习方法，从而推进植物遗传相关机制的研究和作物性状改良。

通讯作者：柳平增 , Email:pzliu@sdau.edu.cn 颜　君 , Email:xinsinian2006@163.com

Current research status of machine learning methods on predicting plant genomic information

ZHANG Zhao-Xu¹ , GUAN Si-Tong¹ , LIN Yi-Ming¹ , SU Pei-Sen² , HUANG Si-Luo³ , MENG Xian-Yong¹ , LIU Ping-Zeng^1,* , YAN Jun^1,*

¹Key Laboratory of Huang-Huai-Hai Smart Agricultural Technology of the Ministry of Agriculture and Rural Affairs, College of Information Science and Engineering, Shandong Agricultural University, Taian 271018, China ²College of Agronomy and Agricultural Engineering, Liaocheng University, Liaocheng 252000, China ³College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China

Abstract:

With the rapid development of high-throughput sequencing technology, massive multi-omics data has been accumulated in plant genomics research. Therefore, how to develop and improve the relevant processing software tools, so as to effectively use these massive data to explore useful biological information, has become an important scientific problem that needs to be solved. Among them, machine learning method has been widely concerned in this field because of its remarkable ability of prediction, classification, data mining and integration. We reviewed the research progresses of machine learning in plant genome on function prediction, focusing on the application results of machine learning model in plant molecular interaction prediction, important functional site prediction, functional annotation, crop breeding and so on. We also look forward to the future development directions and application prospects in this field. This paper will help for plant researchers to quickly understand and apply machine learning methods, so as to contribute to the studies of plant genetic mechanism research and the improvement of crop traits.

Communication Author：LIU Ping-Zeng , Email:pzliu@sdau.edu.cn YAN Jun , Email:xinsinian2006@163.com