基于机器学习的玉米自交系杂种优势类群研究

doi:10.3864/j.issn.0578-1752.2025.02.001

中国农业科学 ›› 2025, Vol. 58 ›› Issue (2): 203-213.doi: 10.3864/j.issn.0578-1752.2025.02.001

• 作物遗传育种·种质资源·分子遗传学 • 上一篇下一篇

基于机器学习的玉米自交系杂种优势类群研究

曹士亮¹(), 张建国¹, 于滔¹, 杨耿斌¹, 李文跃¹, 马雪娜¹, 孙艳杰⁴, 韩微波², 唐贵³, 单大鹏⁴

¹ 黑龙江省农业科学院玉米研究所/玉米国家工程实验室（哈尔滨），哈尔滨 150086
² 黑龙江省农业科学院草业研究所，哈尔滨 150086
³ 黑龙江省农业科学院乡村振兴研究所，哈尔滨 150086
⁴ 黑龙江省农业科学院绥化分院，黑龙江绥化 152052

收稿日期:2024-06-18 接受日期:2024-09-14 出版日期:2025-01-21 发布日期:2025-01-21
联系方式: 曹士亮，Tel：13946013375；E-mail：caoshiliang2003@126.com
基金资助:
黑龙江省省属科研院所科研业务费(CZKYF2023-1-A003); 黑龙江省种业创新与发展项目

Heterosis Groups Research in Maize Inbred Lines Based on Machine Learning

CAO ShiLiang¹(), ZHANG JianGuo¹, YU Tao¹, YANG GengBin¹, LI WenYue¹, MA XueNa¹, SUN YanJie⁴, HAN WeiBo², TANG Gui³, SHAN DaPeng⁴

¹ Maize Institute, Heilongjiang Academy of Agricultural Sciences/National Engineering Laboratory for Maize (Harbin), Harbin 150086
² Institue of Forage and Grassland Sciences, Heilongjiang Academy of Agricultural Sciences, Harbin 150086
³ Rural Revitalization Institute, Heilongjiang Academy of Agricultural Sciences, Harbin 150086
⁴ Suihua Branch of Heilongjiang Academy of Agricultural Sciences, Suihua 152052, Heilongjiang

Received:2024-06-18 Accepted:2024-09-14 Published:2025-01-21 Online:2025-01-21

摘要/Abstract

摘要：

【目的】优化玉米杂种优势类群划分与判别分析方法，为玉米育种提供指导和参考。【方法】采用固相芯片对60份糯玉米自交系进行基因分型，通过质量控制获得不同密度的SNP标记，采用群体结构分析和遗传距离聚类的方法对60份糯玉米进行类群划分，比较不同密度分子标记和分群方法的差异。在此基础上，分别采用随机森林和支持向量机对类群划分结果进行抽样和交叉验证，比较玉米自交系类群判别的预测精度。【结果】通过不同的质量控制标准分别获得11 431和4 022个分子标记，基于2种分子标记密度，分别将60份材料分成5个类群和4个类群，其中，以11 431个SNP标记为基础，通过群体结构分析和遗传距离聚类结果发现，类群内样本一致性为63.33%，以4 022个SNP标记进行分群，发现2种类群划分方法的群内样本一致性为90.00%；比较玉米自交系类群判别的预测精度结果为：基于4 022个标记，随机森林和支持向量机预测精度的平均值（91.43%）高于11 431个标记的预测精度的平均值（86.25%），其中，预测精度最高的是采用4 022个标记的随机森林预测，预测精度为94.17%。【结论】聚类分析法最终将60份玉米糯自交系分为4个类群，运用随机森林和支持向量机对类群划分结果进行抽样和交叉验证，发现随机森林法比支持向量机法能获得更高的预测精度。

关键词: 玉米, 机器学习, 交叉验证, 类群划分, 判别分析

Abstract:

【Objective】The objective of this study is to optimize the classification and discriminant method of maize heterotic groups, and provide guidance and reference for maize breeding practices.【Method】Solid-phase chips were used to genotype 60 waxy maize inbred lines, and high-quality SNP markers with different density were obtained through quality control. Population structure analysis and genetic distance clustering were used to classify the 60 waxy maize inbred lines into different groups, and the differences between different classification methods were compared. On this basis, random forest and support vector machine methods were used to sample and discriminate the results of different classification methods. Five-fold cross-validation was used for sampling, and the prediction accuracy of maize group classification based on different classification methods was compared.【Result】Using different quality control standards, 11 431 and 4 022 molecular markers were obtained, respectively. Based on these two molecular marker densities, 60 materials were divided into 5 and 4 clusters, respectively. When using 11 431 SNP markers, the population structure analysis and genetic distance clustering results showed that the intra-cluster sample consistency was 63.33%. When using 4 022 SNP markers for clustering, the intra-cluster sample consistency was 90.00%. The prediction accuracy results for discriminating maize inbred line clusters showed that the average prediction accuracy (91.43%) of Random Forest and Support Vector Machine using 4 022 markers were higher than that of 11 431 markers (86.25%). Among them, the highest prediction accuracy was achieved by Random Forest using 4 022 markers, with a prediction accuracy of 94.17%.【Conclusion】Clustering analysis ultimately divided 60 waxy maize inbred lines into 4 clusters. Sampling and cross-validation results using Random Forest and Support Vector Machine for cluster classification showed that Random Forest achieved higher prediction accuracy than Support Vector Machine.

Key words: maize, machine learning, cross-validation, clustering, discrimination analysis

曹士亮, 张建国, 于滔, 杨耿斌, 李文跃, 马雪娜, 孙艳杰, 韩微波, 唐贵, 单大鹏. 基于机器学习的玉米自交系杂种优势类群研究[J]. 中国农业科学, 2025, 58(2): 203-213.

CAO ShiLiang, ZHANG JianGuo, YU Tao, YANG GengBin, LI WenYue, MA XueNa, SUN YanJie, HAN WeiBo, TANG Gui, SHAN DaPeng. Heterosis Groups Research in Maize Inbred Lines Based on Machine Learning[J]. Scientia Agricultura Sinica, 2025, 58(2): 203-213.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://www.chinaagrisci.com/CN/10.3864/j.issn.0578-1752.2025.02.001

https://www.chinaagrisci.com/CN/Y2025/V58/I2/203

图/表 9

图1

图2

图3

表1

图4

图5

表2

表3

图6

参考文献 25

[1]	MELCHINGER A E, LEE M, LAMKEY K R, HALLAUER A R, WOODMAN W L. Genetic diversity for restriction fragment length polymorphisms and heterosis for two diallel sets of maize inbreds. Theoretical and Applied Genetics, 1990, 80(4): 488-496. doi: 10.1007/BF00226750 pmid: 24221007
[2]	刘志斋, 吴迅, 刘海利, 李永祥, 李清超, 王凤格, 石云素, 宋燕春, 宋伟彬, 赵久然, 赖锦盛, 黎裕, 王天宇. 基于40个核心SSR标记揭示的820份中国玉米重要自交系的遗传多样性与群体结构. 中国农业科学, 2012, 45(11): 2107-2138. doi: 10.3864/j.issn.0578-1752.2012.11.001.
	LIU Z Z, WU X, LIU H L, LI Y X, LI Q C, WANG F G, SHI Y S, SONG Y C, SONG W B, ZHAO J R, LAI J S, LI Y, WANG T Y. Genetic diversity and population structure of important Chinese maize inbred lines revealed by 40 core simple sequence repeats (SSRs). Scientia Agricultura Sinica, 2012, 45(11): 2107-2138. doi: 10.3864/j.issn.0578-1752.2012.11.001. (in Chinese)
[3]	赵久然, 李春辉, 宋伟, 王元东, 张如养, 王继东, 王凤格, 田红丽, 王蕊. 基于SNP芯片揭示中国玉米育种种质的遗传多样性与群体遗传结构. 中国农业科学, 2018, 51(4): 626-644. doi: 10.3864/j.issn.0578-1752.2018.04.003.
	ZHAO J R, LI C H, SONG W, WANG Y D, ZHANG R Y, WANG J D, WANG F G, TIAN H L, WANG R. Genetic diversity and population structure of important Chinese maize breeding germplasm revealed by SNP-chips. Scientia Agricultura Sinica, 2018, 51(4): 626-644. doi: 10.3864/j.issn.0578-1752.2018.04.003. (in Chinese)
[4]	杨亚桐, 董安忆, 刘松涛, Zenda Tinashe, 段会军. 基于SSR分子标记的糯玉米遗传多样性研究. 江苏农业科学, 2020, 48(2): 83-86.
	YANG Y T, DONG A Y, LIU S T, TINASHE Z, DUAN H J. Study on genetic diversity of waxy corn based on simple sequence repeats (SSR) molecular markers. Jiangsu Agricultural Sciences, 2020, 48(2): 83-86. (in Chinese)
[5]	卢柏山, 史亚兴, 宋伟, 徐丽, 赵久然. 利用SNP标记划分甜玉米自交系的杂种优势类群. 玉米科学, 2015, 23(1):58-62, 68.
	LU B S, SHI Y X, SONG W, XU L, ZHAO J R. Heterotic grouping of sweet corn inbred lines by SNP markers. Journal of Maize Sciences, 2015, 23(1): 58-62, 68. (in Chinese)
[6]	徐磊, 徐志军, 安东升, 胡小文, 高玉尧, 刘洋. 基于SNP标记的糯玉米指纹图谱构建和遗传多样性分析. 分子植物育种, 2022, 20(19):6405-6414.
	XU L, XU Z J, AN D S, HU X W, GAO Y Y, LIU Y. Fingerprints construction and genetic diversity analysis of waxy corns based on SNP markers. Molecular Plant Breeding, 2022, 20(19): 6405-6414. (in Chinese)
[7]	卢媛, 艾为大, 韩晴, 王义发, 李宏杨, 瞿玉玑, 施标, 沈雪芳. 糯玉米自交系SSR标记遗传多样性及群体遗传结构分析. 作物学报, 2019, 45(2):214-224. doi: 10.3724/SP.J.1006.2019.83008
	LU Y, AI W D, HAN Q, WANG Y F, LI H Y, QU Y J, SHI B, SHEN X F. Genetic diversity and population structure analysis by SSR markers in waxy maize. Acta Agronomica Sinica, 2019, 45(2): 214-224. (in Chinese)
[8]	李扬. 机器学习在农业领域应用现状与前景. 安徽农学通报, 2021, 27(1):164-165, 174.
	LI Y. Application status and prospect of machine learning in agriculture. Anhui Agricultural Science Bulletin, 2021, 27(1): 164-165, 174. (in Chinese)
[9]	薛菁菁, 陈慧敏, 孔令怡, 樊欣怡, 聂飞平. 机器学习的基石: 聚类任务的现状与挑战. 科学观察, 2024, 19(1):4-17. doi: 10.15978/j.cnki.1673-5668.202401002
	XUE J J, CHEN H M, KONG L Y, FAN X Y, NIE F P. The foundation of machine learning: current status and challenges in clustering tasks. Science Focus, 2024, 19(1): 4-17. (in Chinese)
[10]	BRADBURY P J, ZHANG Z W, KROON D E, CASSTEVENS T M, RAMDOSS Y, BUCKLER E S. TASSEL: Software for association mapping of complex traits in diverse samples. Bioinformatics, 2007, 23(19): 2633-2635. doi: 10.1093/bioinformatics/btm308 pmid: 17586829
[11]	ALEXANDER D H, NOVEMBRE J, LANGE K. Fast model-based estimation of ancestry in unrelated individuals. Genome Research, 2009, 19(9): 1655-1664. doi: 10.1101/gr.094052.109 pmid: 19648217
[12]	JOMBART T. Adegenet: A R package for the multivariate analysis of genetic markers. Bioinformatics, 2008, 24(11): 1403-1405. doi: 10.1093/bioinformatics/btn129 pmid: 18397895
[13]	TEAM R. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2014.
[14]	YU G C, SMITH D K, ZHU H C, GUAN Y, LAM T T Y. Ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods in Ecology and Evolution, 2017, 8(1): 28-36.
[15]	LIAW A, WIENER M. Classification and regression by randomForest. R News, 2002, 2: 18-22.
[16]	KUHN M. Building predictive models in r using the caret package. Journal of Statistical Software, 2008, 28(5): 1-26.
[17]	李念念, 王义波, 徐国平, 易黎, 王爱方, 李婷, 曹刚强. 基于SNP标记的玉米自交系类群划分方法和分群功效评估指标的比较. 植物遗传资源学报, 2020, 21(3):605-618. doi: 10.13430/j.cnki.jpgr.20190618003
	LI N N, WANG Y B, XU G P, YI L, WANG A F, LI T, CAO G Q. Comparison of different grouping procedures and evaluation criteria for grouping maize inbreds using SNP data. Journal of Plant Genetic Resources, 2020, 21(3): 605-618. (in Chinese)
[18]	钟昌松, 徐利远, 余桂蓉, 杜文平, 蒲志刚, 李琳. 20份特用玉米自交系亲缘关系的SSR标记研究. 玉米科学, 2006, 14(4):43-46.
	ZHONG C S, XU L Y, YU G R, DU W P, PU Z G, LI L. Genetic relationship among 20 special maize inbreds studied by SSR markers. Journal of Maize Sciences, 2006, 14(4): 43-46. (in Chinese)
[19]	赵文明, 王森, 陈艳萍, 张美景, 袁建华. 基于60个核心SSR标记的糯玉米自交系遗传多样性分析. 江西农业学报, 2018, 30(12):1-8.
	ZHAO W M, WANG S, CHEN Y P, ZHANG M J, YUAN J H. Genetic diversity analysis of waxy corn inbred lines based on 60 core SSR markers. Acta Agriculturae Jiangxi, 2018, 30(12): 1-8. (in Chinese)
[20]	史亚兴, 卢柏山, 宋伟, 徐丽, 赵久然. 基于SNP标记技术的糯玉米种质遗传多样性分析. 华北农学报, 2015, 30(3): 77-82. doi: 10.7668/hbnxb.2015.03.015
	SHI Y X, LU B S, SONG W, XU L, ZHAO J R. Acta Agriculturae Boreali-Sinica, 2015, 30(3): 77-82. (in Chinese)
[21]	孙佩, 张培风, 张玉红, 周联东, 王蕊, 王文洁, 张瑞平, 李祥, 马朝阳, 李合顺, 王学军. 基于SNP标记的玉米自交系S155、PHA458和A01的遗传多样性分析. 陕西农业科学, 2023, 69(9):1-7.
	SUN P, ZHANG P F, ZHANG Y H, ZHOU L D, WANG R, WANG W J, ZHANG R P, LI X, MA C Y, LI H S, WANG X J. Genetic diversity analysis of maize inbred lines S155, PHA458 and A01 based on SNP markers. Shaanxi Journal of Agricultural Sciences, 2023, 69(9): 1-7. (in Chinese)
[22]	肖颖妮, 于永涛, 谢利华, 祁喜涛, 李春艳, 文天祥, 李高科, 胡建广. 基于SNP标记揭示中国鲜食玉米品种的遗传多样性. 作物学报, 2022, 48(6):1301-1311. doi: 10.3724/SP.J.1006.2022.13031
	XIAO Y N, YU Y T, XIE L H, QI X T, LI C Y, WEN T X, LI G K, HU J G. Genetic diversity analysis of Chinese fresh corn hybrids using SNP Chips. Acta Agronomica Sinica, 2022, 48(6): 1301-1311. (in Chinese)
[23]	李余良, 索海翠, 韩福光, 刘建华, 胡建广, 高磊, 李武. 基于SLAF-seq技术分析甜、糯玉米种质遗传多样性. 玉米科学, 2019, 27(4):71-78.
	LI Y L, SUO H C, HAN F G, LIU J H, HU J G, GAO L, LI W. Analysis of genetic diversity of sweet and wax corn germplasms using SLAF-seq technology. Journal of Maize Sciences, 2019, 27(4): 71-78. (in Chinese)
[24]	王欣, 徐一亿, 徐扬, 徐辰武. 作物全基因组选择育种技术研究进展. 生物技术通报, 2024, 40(3):1-13. doi: 10.13560/j.cnki.biotech.bull.1985.2023-1079
	WANG X, XU Y Y, XU Y, XU C W. Research progress in genomic selection breeding technology for crops. Biotechnology Bulletin, 2024, 40(3): 1-13. (in Chinese)
[25]	YIN L L, ZHANG H H, ZHOU X, YUAN X H, ZHAO S H, LI X Y, LIU X L. KAML: improving genomic prediction accuracy of complex traits using machine learning determined parameters. Genome Biology, 2020, 21(1): 146. doi: 10.1186/s13059-020-02052-w pmid: 32552725

分群方法 Clustering method	A群 Group A	B群 Group B	C群 Group C	D群 Group D	E群 Group E
群体结构分群 STR	N1, N2, N3, N4, N5, N6, N7, N8, N9, N10, N11, N12, N13, N15, N16, N32	N38, N39, N40, N41, N42, N43, N44	N14, N18, N19, N20, N21, N22, N23, N24, N25, N26	N17, N28, N29, N30, N31, N33, N34, N35, N36, N37, N48, N50, N52, N53, N54, N55, N56, N57, N58, N60	N27, N45, N46, N47, N49, N51, N59
遗传距离分群 GD	N1, N2, N3, N4, N5, N6, N7, N8, N9, N10, N11, N12, N14, N16, N17, N32, N34	N38, N39, N40, N41, N42, N43, N44, N48, N50	N18, N19, N20, N21, N22, N23, N24, N25, N26, N27, N29, N31, N35, N37, N49, N51, N54, N59, N60	N13, N15, N28, N30, N33, N36, N45, N46, N47, N52, N53, N55, N56	N57, N58

分群方法 Clustering method	A群 Group A	B群 Group B	C群 Group C	D群 Group D
群体结构分群 STR	N1, N2, N3, N4, N5, N6, N7, N8, N9, N10, N11, N12, N16, N32, N13, N15, N14, N17	N38, N39, N40, N41, N42, N43, N44	N18, N19, N20, N21, N22, N23, N24, N25, N26, N29, N34	N45, N46, N47, N59, N51, N53, N27, N28, N30, N49, N56, N37, N31, N48, N54, N55, N52, N33, N50, N35, N36, N60, N57, N58
遗传距离分群 GD	N1, N2, N3, N4, N5, N6, N7, N8, N9, N10, N11, N12, N16, N32, N13, N15, N14, N17	N38, N39, N40, N41, N42, N43, N44, N48, N50, N55, N57, N58	N18, N19, N20, N21, N22, N23, N24, N25, N29, N34	N26, N27, N28, N30, N31, N33, N35, N36, N37, N45, N46, N47, N49, N51, N52, N53, N54, N56, N59, N60

分群方法 Grouping methods	预测方法 Prediction methods	标记数 Marker number		预测方法均值 Mean of prediction method (%)	分群方法均值 Mean of grouping methods (%)
分群方法 Grouping methods	预测方法 Prediction methods	11431	4022	预测方法均值 Mean of prediction method (%)	分群方法均值 Mean of grouping methods (%)
群体结构 STR	随机森林RF	87.50%/0.11	94.17%/0.07	90.84	89.83
群体结构 STR	支持向量机SVM	87.22%/0.11	90.42%/0.10	88.82	89.83
遗传距离 GD	随机森林RF	86.25%/0.11	94.17%/0.06	90.21	87.40
遗传距离 GD	支持向量机SVM	84.03%/0.13	86.94%/0.10	85.49	87.40
	均值Mean (%)	86.25	91.43

基于机器学习的玉米自交系杂种优势类群研究

Heterosis Groups Research in Maize Inbred Lines Based on Machine Learning

RichHTML

PDF

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 25

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	张斯佳, 杨杰, 赵帅, 李莉威, 王贵彦. 华北平原多样化作物与小麦-玉米轮作对土壤质量的影响[J]. 中国农业科学, 2025, 58(2): 238-251.
[2]	刘静, 汪泓, 张磊, 肖玖军, 吴建高, 龚明冲. 基于高光谱的辣椒叶片氮素含量反演[J]. 中国农业科学, 2025, 58(2): 252-265.
[3]	曹言勇, 程泽强, 马娟, 杨文博, 朱卫红, 孙新艳, 李慧敏, 夏来坤, 段灿星. 整合转录组和代谢组学分析揭示玉米对层出镰孢茎腐病的响应机制[J]. 中国农业科学, 2025, 58(1): 75-90.
[4]	吕金岭, 尤克, 王小非, 肖强, 李文峰, 马进, 杨清, 张金平, 孔海江, 常运华. 河南省典型农区近地表大气氨时空分布特征及主要影响因素[J]. 中国农业科学, 2025, 58(1): 127-140.
[5]	范虹, 殷文, 胡发龙, 樊志龙, 赵财, 于爱忠, 何蔚, 孙亚丽, 王凤, 柴强. 绿洲灌区密植对氮肥减量玉米产量的补偿潜力[J]. 中国农业科学, 2024, 57(9): 1709-1721.
[6]	王程泽, 张燕, 付伟, 贾京哲, 董金皋, 申珅, 郝志敏. 玉米ACO基因家族生物信息学及表达模式分析[J]. 中国农业科学, 2024, 57(7): 1308-1318.
[7]	倪书辉, 史东梅, 盘礼东, 叶青, 伍俊豪. 紫色土坡耕地耕层持水抗旱性能及生产力对侵蚀程度的响应[J]. 中国农业科学, 2024, 57(7): 1350-1362.
[8]	高晨曦, 郝陆洋, 胡悦, 李永祥, 张登峰, 李春辉, 宋燕春, 石云素, 王天宇, 黎裕, 刘旭洋. 干旱条件下玉米转座子插入关联的表观调控分析[J]. 中国农业科学, 2024, 57(6): 1034-1048.
[9]	张博文, 赵丽雯, 徐璐, 李盼, 曾凡力, 孟亚南, 董金皋. 玉米大斑病菌细胞周期蛋白依赖性激酶结构亚基StCks1鉴定及其基因功能分析[J]. 中国农业科学, 2024, 57(5): 900-908.
[10]	王语, 张渝鹏, 朱冠亚, 廖航烯, 侯文峰, 高强, 王寅. 局部供氮对干旱胁迫下玉米苗期生长发育和水氮利用的影响[J]. 中国农业科学, 2024, 57(5): 919-934.
[11]	高尚洁, 刘杏认, 李迎春, 柳晓婉. 施用生物炭和秸秆还田对农田温室气体排放及增温潜势的影响[J]. 中国农业科学, 2024, 57(5): 935-949.
[12]	李乾川, 许世卫, 张永恩, 庄家煜, 李灯华, 刘保花, 朱之洵, 刘浩. 基于气象因素的玉米单产堆栈集成学习建模与预测[J]. 中国农业科学, 2024, 57(4): 679-697.
[13]	石德杨, 李艳红, 王飞飞, 夏德君, 矫岩林, 孙妮娜, 赵健. 高密度下扩行缩株对夏玉米干物质与养分积累、转运的调控效应[J]. 中国农业科学, 2024, 57(23): 4658-4672.
[14]	罗钦, 毛方华, 陈谢勇, 陈静, 黄彪, 叶乃兴, 郑德勇, 韦航, 姚清华. 不同生境白牡丹茶叶香气特征分析及产地判别[J]. 中国农业科学, 2024, 57(23): 4774-4793.
[15]	曹文茁, 于振文, 张永丽, 张振, 石玉, 王永军. 不同施氮量下黑土地春玉米籽粒淀粉积累动态及产量形成差异[J]. 中国农业科学, 2024, 57(22): 4431-4443.