The choice of parents to cross is the most crucial step in any crop improvement
program. A better understanding of genetic diversity among the available parental
genotypes would help the breeder to make better crosses. Several molecular marker
techniques have been used to assess genetic diversity in sugarcane. Target Region
Amplification Polymorphism (TRAP) is fairly a new marker technique which is believed
to amplify genic regions of the genome. In this study, TRAP, AFLP, and pedigree data
were used to estimate genetic similarity (GS) among nine sugarcane genotypes often used
as parents. The objectives were to compare TRAP with AFLP and pedigree-based
coefficient of parentage (COP) for estimating GS among sugarcane genotypes. Twelve
TRAP primer combinations produced a total of 444 bands, out of which 242 (55 %) were
polymorphic, whereas 28 AFLP primer combinations produced a total 1325 bands out of
which 686 (53 %) were polymorphic. TRAP-based GS estimates ranged from 0.67 to
0.87 with a mean of 0.75, while AFLP-based estimates ranged from 0.72 to 0.84 with a
mean of 0.76. The COP-based GS estimates ranged from 0.03 to 0.36 with a mean of
0.12. The dendrograms were constructed using the unweighted pair-group method with
arithmetic mean (UPGMA). Although no distinct pattern was observed in the COP
dendrogram, the TRAP dendrogram was better explained by the pedigree records. The
AFLP dendrogram showed some distinct cluster patterns. The associations between
TRAP-COP (r = 0.41) and AFLP-COP (r = 0.42) were moderate, whereas the TRAPAFLP
(r = 0.14) association was low. Our results indicate that all the three methods
estimate different aspects of GS. Therefore, based on the objectives of the research, some
combination of TRAP, AFLP and COP would be a better choice in making decisions of
which parents to cross in a crop improvement program.