摘要
为了高效保护、管理以及利用大黄鱼种质资源,迫切需要开发精准的大黄鱼遗传种质鉴定技术。基于前期开发的大黄鱼“宁芯3号”55K液相基因分型芯片,本研究对中国沿海野生群体、闽浙养殖群体和多个选育系共计21个大黄鱼群体进行遗传种质鉴定。群体遗传学分析结果揭示大黄鱼群体可划分为南海群体、闽东群体和岱衢群体,其中南海群体遗传分化最为显著。基于机器学习的大黄鱼群体分类结果显示,未知大黄鱼个体所属地理群体鉴定准确率大于99%。未知大黄鱼个体所属遗传改良选育系也具有极高的鉴定准确率,例如经过3代选育的抗刺激隐核虫新品系GS3F3,基于神经网络的鉴定精确率可以达到99%。研究表明,利用“宁芯3号”芯片和机器学习方法可快速实现大黄鱼种质的精准鉴定。本研究为大黄鱼种质资源精准鉴定和种质管理、育种材料和新品种知识产权保护等提供了有效的工具和解决方案,也可为其他水产生物种质资源鉴定提供借鉴。
Larimichthys crocea is an important commercial fish in China,with an annual production of more than 250,000 tons in recent years.L.crocea is extremely rich in genetic resources,which consisted of wild populations distributed in natural sea areas and breeding lines obtained through decades of selection breeding.There is an urgent need to develop an accurate genetic identification method to distinguish different germplasm of L.crocea to efficiently protect,manage and utilize L.crocea genetic resources.However,the lack of high-throughput genotyping tools for L.crocea and the lack of representative samples of geographical populations have made accurate identifiing genetic resources difficult.Based on the previously developed 55K liquid SNP array("Ningxin III")for L.crocea,the present study aims to carry out genetic identification for 21 L.crocea populations,including wild populations in coastal China,cultured populations in Fujian and Zhejiang,and multiple breeding lines.The results of population genetic analysis revealed that the L.crocea population could be divided into Nanhai,Mindong and Daiqu populations,among which the genetic differentiation of the Nanhai population was the most significant.The classification results of large yellow croaker populations based on machine learning methods showed that the identification accuracy rate of the geographical group to which unknown L.crocea individuals belong was more than 99%.The breeding lines to which unknown L.crocea individuals belong also had a very high identification accuracy rate.For example,after three generations of genetic selection,a new strain(GS3F3)that had strong resistance against Cryptocaryon irritans had an identification accuracy rate of 99%based on the neural network method.The present research showed that"Ningxin III"chip and machine learning methods could be used to implement quick and accurate genetic identification for L.crocea.The present study provided an effective tool for accurately identifying and managing the genetic resources of L.crocea,intellectual property protection for breeding materials and lines.Also,it provided a reference for the genetic identification of other aquatic organisms.In the future,it is necessary to establish a complete database covering all L.crocea germplasm resources and genetic identification standards,and develop a supporting visual computer program to perform identification work.
作者
赵吉
冯苗胜
柯巧珍
王家迎
江汀森
吴雄飞
彭士明
白玉麟
沈伟良
周涛
濮菲
徐鹏
ZHAO Ji;FENG Miaosheng;KE Qiaozhen;WANG Jiaying;JIANG Tingsen;WU Xiongfei;PENG Shiming;BAI Yulin;SHEN Weiliang;ZHOU Tao;PU Fei;XU Peng(State Key Laboratory of Mariculture Breeding,College of Ocean and Earth Sciences,Xiamen University,Xiamen 361101,China;State Key Laboratory of Marine Environmental Science,College of the Environment and Ecology,Xiamen University,Xiamen 361101,China;Ningbo Academy Oceanology and Fishery,Ningbo 315048,China;East China Sea Fisheries Research Institute,Chinese Academy Fishery Sciences,Shanghai 200090,China)
出处
《水产学报》
CAS
CSCD
北大核心
2024年第12期28-39,共12页
Journal of Fisheries of China
基金
国家重点研发计划(2022YFD2401002)
国家杰出青年科学基金(32225049)
福建省种业创新与产业化项目(2021FJSCZY01)。
关键词
大黄鱼
宁芯3号
种质鉴定
机器学习
Larimichthys crocea
"Ningxin III"chip
germplasm identification
machine learning
作者简介
第一作者:赵吉,从事大黄鱼遗传选育研究,E-mail:zhaoji@xmu.edu.cn;通信作者:徐鹏,从事鱼类遗传学与遗传育种研究,E-mail:xupeng77@xmu.edu.cn。
