Machine Learning-Based Heading Date QTL Detection in Rice

• 영문명: Machine Learning-Based Heading Date QTL Detection in Rice
• 발행기관: 한국육종학회
• 저자명: Seung Young Lee Jae-Hyuk Han Hyeok-Jin Bak Su-Kyung Ha Hyun-Sook Lee Gileung Lee Jae-Ryoung Park Kyeongmin Kang Jung-Pil Suh Mina Jin Ji-Ung Jeung Youngjun Mo
• 간행물 정보: 『Plant breeding and biotechnology』Vol.13, 108~118쪽, 전체 11쪽
• 주제분류: 농수해양 > 기타농수해양
• 파일형태: PDF
• 발행일자: 2025.02.28

국문 초록

Abstract Quantitative trait locus (QTL) analysis is a powerful approach for identifying variantsassociated with the phenotypic variation of complex traits. However, selecting optimalmethods and pre-processing steps require considerable time and effort. In this study,we demonstrated applicability and replicability of machine learning (ML) models in QTLanalysis by evaluating their performance in comparison with conventional QTL analysismethods using 142 recombinant inbred lines derived from two japonica rice cultivars,Koshihikari and Baegilmi. Random forest and gradient boosting models showed the highestpredictive accuracy, and consistently identified three QTLs associated with headingdate: qDTH3, qDTH6, and qDTH7. Moreover, ML-based QTL analysis detected minor-effectqDTH10, where Koshihikari allele promoted heading date when combined withKoshihikari alleles of qDTH6 and qDTH7. These results demonstrate the applicability of MLmodels in QTL analysis on bi-parental mapping population in rice.

키워드

Machine learning QTL Rice Heading date