文摘
Genetic relationships among 42 cultivars of Canna were determined by using amplified fragment length polymorphism (AFLP) marker. A total of 1607 DNA fragments was produced with 25 AFLP primer combinations and out of which 1491 (92.78 % ) were found polymorphic and 116 (7.22 % ) monomorphic. The number of polymorphic fragments varied from 33 (E-ACA/M-CAA) to 86 (E-ACT/M-CAA) with an average of 59.6 per primer combination and percent polymorphism varied from 81.7 % (E-AAG/M-CAA) to 100 % (E-ACC/M-CTT) with an average of 92.8 % per primer combination. The polymorphism information content (PIC) value ranged from 0.24 to 0.35 with an average of 0.30 per fragment and the highest PIC value (0.35) was noticed for primer combination E-AAG/M-CAC followed by E-ACT/M-CTA, E-ACA/M-CAA, E-AAG/M-CAA (0.34). Marker index (MI) and resolving power (RP) varied from 11.22 to 26.66 and 17.10 to 40.00 respectively. Jaccard's similarity coefficient varied from 0.33 to 0.72 with an average of 0.49 ¡À 0.03. The maximum genetic similarities (72 % ) were noticed between the cultivar NBC_1 and NBC_2; NBC_16 and NBC_19 followed by NBC_19 and NBC_30 (70 % ). Based upon genetic similarity coefficient the cultivars NBC_43, NBC_24, NBC_38, NBC_22, NBC_29 and NBC_36 were found to be most divergent among all the cultivars. The UPGMA clustering revealed four major groups accommodating 93 % cultivars and three cultivars each of different species i.e. C. argentina (NBC_24), C. latifolia (NBC_43) and C. generalis (NBC_13) did not grouped with any clusters. Model based clustering obtained from STRUCTURE analysis also reveals similar pattern of grouping.