Molecular characterization and genetic relationships among cotton genotypes 2- AFLP Analysis

The high-resolution genotyping method of amplified fragment length polymorphism (AFLP) was used to study the genetic relationships among 21 cotton genotypes from two different species G. barbadense and G. hirsutum. Sixteen AFLP primer combinations were used to selectively amplify the DNA fragments that matches the primer-extension sequence to investigate the genetic polymorphism among the 21 cotton genotypes. The 16 AFLP primer combinations produced 940 bands among which 474 were polymorphic, thus, representing a level of polymorphism of 50.4% among the 21 cotton genotypes. The amplification of AFLP templates resulted in a number of reproducible fragments ranging from 31 to 90 per primer with a size range of 60 bp to 780 bp. Fifteen primer combinations detected unique specific markers identifying 8 out of the 21 genotypes. A dendrogram was generated from the AFLP information that revealed two main clusters. All the genotypes belonging to G. barbadense except one (Pima Early American) were grouped in one cluster, while the accessions representing G. hirsutum constituted the second cluster, thus, confirming the results previously obtained by RAPD, ISSR and SSR analysis on the same cotton genotypes. To evaluate the efficiency of the different marker systems, the sum effective number of alleles (SENA), the average expected heterozygosity for polymorphic markers (Hav(p)), the effective multiplex ratio (E) and marker index (MI) were calculated. The AFLP exhibited considerably high SENA (318.2) compared to RAPD, ISSR and SSR (127.7, 46.0 and 22.3, respectively). The average heterozygosity values were comparable for the different marker systems (0.39, 0.36, 0.39 and 0.34 in AFLP, RAPD, ISSR and SSR, respectively). The MI was 182.2 in AFLP’s, while it was 73.6, 26.3 and 12.7 in RAPD, ISSR and SSR, respectively. Thus, the results indicated that AFLP is more effective in detecting high level of polymorphism. The correlation coefficient was considerably higher between SSR and ISSR (0.61), and it was lower between RAPD’s and AFLP’s (0.26) than that between AFLP and ISSR (0.44) and AFLP and SSR (0.49). The results confirmed that different marker systems differ in the mechanism of detecting polymorphism, genome coverage and the ease of application. Therefore, they could complement each other to draw more accurate conclusions.