Abstract
Erosion of genetic diversity is a side effect of widely used selection-based plant breeding schemes which ultimately ensued to cultivars bearing a narrow genetic base. The present study was carried out at Cotton Research Station, Vehari, Pakistan during the year 2020-21. The aim was to explore the genetic diversity present in the studied genotypes. Principal component analysis (PCA), correlation study, K- mean clustering, and biplot analysis techniques were applied to 102 upland cotton genotypes. Sowing was completed by following a randomized complete block design (RCBD) with three replications. The first three principal components (PC) depicted eigenvalue >1 and enfolded with 53.159% of aggregate variability. Seed cotton yield, ginning out turn (GOT), staple length, and micronaire value (MV) were loaded to PC-1. The plant stature was loaded to PC-2, while the plant population indicated relatedness to PC-3. The MV revealed a positive correlation (0.341) with GOT, while negative bonding (-0.204) with staple length (p<0.05). Cluster analysis classified the studied genotypes into 7 clusters. The highest Mahalanobis distance (16828) was noted between genotypes located in clusters 2 and 6. Biplot analysis demonstrated the unique scatter behaviour of GP-1, GP-8, GP-21, GP-38, GP-39 and VH-259 (check) genotypes being located at the vertex of the polygon. The genotypes in clusters 2 and 6 might be used as parents in future gene pyramiding schemes to breed cotton cultivars with a broad genetic base. Such imminent cultivars will be able to combat incipient stress factors due to climate change.