Identifying superior finger millet ( <i>Eleusine coracana</i> L.) landraces using the multi-trait genotype-ideotype distance index (MGIDI)

Genetic variability and selection are crucial for finger millet breeding programs to maximize genetic gain and productivity. This study evaluated 18 finger millet landraces cultivated in the mid-hills of Nepal, along with ‘Kavre Kodo 2’ as a check variety, assessing 14 agro-morphological traits. The objective was to identify superior finger millet landraces using a multivariate approach based on the multi-trait genotype–ideotype distance index (MGIDI). The experiment was conducted in a randomized complete block design (RCBD) with three replications. Principal component analysis revealed that the first five components collectively accounted for 80.8% of the total variation, reflecting the genetic relatedness among the landraces. Broad-sense heritability (h2) ranged from 40.4% to 97.5%, with most traits exceeding 60%, indicating a strong genetic influence and promising prospects for effective selection. MGIDI identified ‘Seto Kodo’, ‘Larparey’, and ‘Dudhey’ as ideotype landraces with a selection intensity of 15%. These landraces exhibited positive gains across all traits studied, except for flag leaf area and the number of effective tillers. The strengths-weaknesses analysis emphasized their value for targeted improvement. Overall, the findings highlight MGIDI as an effective tool for selecting superior finger millet landraces and guiding breeding strategies to improve productivity.