Selection of parents for hybridization relies on the information on their genetic relationship and diversity which are essential in any breeding program. This study aimed to estimate the extent of genetic diversity and population structure of 76 sugarcane accessions from seven regions in the Philippines using 57 morphological characters and 50 microsatellite markers. The sugarcane collections exhibited moderate to high diversity with mean of H’ = 0.72 for qualitative and H’ = 0.75 for quantitative morphological characters, respectively. This is corroborated by the analysis of variance (ANOVA) of agronomic parameters, except for stalk length. Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on morphology subdivided the accessions into 31 clusters which reveal phenotypic variability among sugarcane samples. The fingerprints of the 76 accessions were also evaluated using 45 Saccharum-based genomic SSR and 5 EST-SSR primer pairs to measure genetic diversity and population structure. Based on UPGMA, a total of six clusters were generated with a 0.65 coefficient of dissimilarity, and the sugarcane accessions were further subdivided into five major sub-populations. Out of 50 markers used, 41 (82%) were found to be highly informative with a mean PIC value of 0.69. It was also expected that Saccharum-based genomic SSRs were more polymorphic (92%) compared to EST-SSRs (82%) since the latter preferably amplify in more conserved and expressed sequences in the genome. Out of 2,850 pairwise combinations based on the Jaccard coefficient index, large diverse parental combinations (genetic dissimilarity= 0.51 - 0.70) were observed, indicating substantial diversity from the existing breeding pool of IPB-UPLB for genetic improvement. Cluster analysis based on UPGMA, STRUCTURE analysis, and Principal Coordinate Analysis (PCoA) were predominantly consistent. However, no association was observed between the geographical origin and genetic distance of the genotypes based on molecular data. The results showed that accessions were grouped into five sub-populations and genetic differentiation within sub-population was higher (85%) as compared to between sub-population (15%) based on the analysis of molecular variance (AMOVA), suggesting an active exchange of genetic pool across provinces and regions of both Luzon and Visayas islands. The findings from this study will be useful for future breeding efforts by exploiting genetic variation existing in the current breeding population.