摘要 : Identification of superior parental lines require phenotypic or genotypic characterization of the germplasm, which is time consuming and laborious. The advanced potato breeding clones including control varieties were evaluated und... 展开 Identification of superior parental lines require phenotypic or genotypic characterization of the germplasm, which is time consuming and laborious. The advanced potato breeding clones including control varieties were evaluated under field conditions to identify the suitable parental lines for tuber yield, late blight and specific gravity based on estimated breeding values using historical data as training population during 2008-09 to 2019-20. Pedigree of all the clones was searched to a depth of 5-6 generations. The mixed model fitting using REML was true. The minimum and maximum values for total tuber yield, late blight area under disease progress curve (AUDPC) and specific gravity were 28.33-732.2, 0-1846.25, 1.043-1.087, respectively. Pedigree had 161 founders out of total 759 individuals and the pedigree matrix showed relationship in 1,09,057 pairs. Heritability estimate was highest for specific gravity (0.77) followed by late blight AUDPC (0.49) and total tuber yield (0.24). The estimate of the BLUP breeding values (EBVs) for total tuber yield were highest for Kufri Mohan, Kufri Pukhraj, Kufri Khyati, SM/14-342 and Kufri Pushkar while SM/10-05, SM/09-99, Kufri Girdhari, SM/11-120 and Kufri Karan for late blight resistance, and Kufri Chipsona-1, SM/14-225, SM/14-229, SM/10-220 and SM/09-153 for specific gravity were the best genotypes. The prediction accuracy measured as correlation between EBV and observed phenotype score for specific gravity (0.97) was near perfect, very high for late blight AUDPC (0.75), and high for total tuber yield (0.55). The prediction bias measured as regression coefficient and scatter plots of observed phenotype values on EBVs was nil to low for specific gravity andlate blight AUDPC, while underestimation was observed for total tuber yield. Overall, the prediction accuracies for all the three traits were very good and selection of parental lines using Best Linear Unbiased Prediction breeding values could result in enhanced genetic gain in potato breeding. 收起