Germplasm collection and genomic DNA extraction A total of 12 rice genotypes were evaluated in this study including Jamaisohagi, Sugandha, Darsail, Chinikani, Bhogganijia, Dolargura, Depa, Patnai-23, Opchaya, Holdijorun, Jingasail and Basmati PNR 346. All seeds those collected from the Genetic Resource Center (GRC), Bangladesh Rice Research Institute (BRRI) were germinated at aseptic condition by keeping them at 30ºC for 1 day and raised in pots in a net house. At 3 weeks of age, about 2 cm of leafs from each plant was harvested and bulked for each genotype. Total genomic DNA was extracted from the bulked leaf samples by following a miniprep DNA extraction protocol, which did not require liquid nitrogen and required only a small amount of tissue samples (Zheng et al., 1995). The quality of DNA was also checked by DNA quantification using a Thermo Scientific NanoDrop™ 1000 spectrophotometer (Thermo Fisher Scientific, USA). All chemicals used for DNA extraction were purchased from Sigma-Aldrich, Germany. SSR markers and PCR amplification Twenty four SSR primer pairs (Promega Corporation, USA) were selected on the basis of the published rice microsatellite framework map for the genetic diversity analysis of the 12 aromatic rice cultivars in Bangladesh. Primers that showed polymorphic banding patterns were selected whereas primers that showed monomorphic banding patterns were excluded. Finally, 9 microsatellite primers with a distinct chromosome number were used for final polymerase chain reaction (PCR) amplification. Information regarding the original source, repeat motifs, primer sequences, expected length, chromosomal localizations and repeat types of the SSRs can be found in the Web database (http://www.gramene.org). Prior to DNA amplification, a PCR cocktail was prepared containing all required components. All reagents were purchased from Sigma Aldrich. PCR amplification reactions were done in 10 l reaction mixtures, containing 3 ?l of diluted template DNA, 0.5 ?l of each forward and reverse primer, 0.25 of 10 mM dNTPs, 1.5 of 10x buffer, 0.2 l of Taq polymerase, 1.8 l of MgCl2 and 2.25 of ddH2O. An DNA thermal cycler (Model: ALS 1296, BioRad, USA and G-STORM, GSI, England, Serial no: GT-11620) was used along with the following PCR profile: an initial denaturation step for 5 min at 94°C (hot start and strand separation), followed by 34 cycles of denaturation (94°C), annealing (55°C) and primer elongation (72°C) for 30 seconds each and then a final extension at 72°C for 5 min. Amplified products were stored at -20°C until further use. Electrophoretic separation and visualization of amplified products Prior to electrophoresis, each PCR product was mixed with gel loading dye (bromophenol blue, xylene cyanol and sucrose) and electrophoresis was carried out in a mini vertical electrophoresis tank (CBS Scientific Co Inc., CA. USA), run on 8% polyacrylamide gels in TBE buffer. Four microliters of the sample were loaded in each well and run at 80 V for 90 minutes. The gel, after electrophoresis, was stained with ethidium bromide for 30-35 min, kept in dark, and then scanned using an UVPRO (Uvipro Platinum, EU) gel documentation unit linked to a PC. The reproducibility of amplification products was confirmed twice for each primer. SSR data analysis The size of most intensely amplified fragments was determined by comparing the migration distance of amplified fragments relative to the molecular weight of known size markers, 50 base pairs (bp) DNA ladder using Alpha-Ease FC 5.0 software (Alpha Innotech, USA). The number of alleles per locus, major allele frequency, gene diversity and PIC values were calculated using Power Marker version 3.25 (Liu & Muse, 2005). All the genotypes were scored for the presence and absence of the SSR bands throughout all 12 genotypes and the data were exported to binary data for the presence (1) or absence (0) or as a missing observation for further analysis with NTSYS-pc version 2.2 (Rohif, 2002). NTSYS-pc was used to construct a UPGMA (unweighted pair group method with arithmetic averages) dendrogram showing the distance-based interrelationship among the genotypes. For the unrooted phylogenetic tree, genetic distance was calculated using the “C.S Chord 1967” distance (Cavalli-Sforza & Edwards, 1967) in Power Marker with tree viewed using Tree view software.