Genetic studies along with characterization of elite breeding lines provide understanding ofrnthe genetic diversity and relationship among the inbred lines. They also offer information onrnthe type of gene action controlling the inheritance of desirable quantitative traits. Therninformation enables breeders to define a systematic breeding strategy and to select suitablernparents and hybrids for further breeding activities and commercialization. Thus, thernpurposes of this study were to estimate combining abilities and heterosis and to determinerngenetic variation, correlation, heritability and expected genetic advance of elite qualityrnprotein maize (QPM) inbred lines for grain yield and other agronomic traits as well as torninvestigate their genetic diversity and relationship using SNP and ISSR markers. A total ofrn116 QPM test cross hybrids developed by crossing 58 QPM inbred lines with two QPMrntesters was evaluated for 17 morphological traits along with two conventional maize (CM)rnand two QPM standard checks in drought stressed areas of eastern Ethiopia â€“ MelkassarnAgricultural Research Center (MARC), Edo Gojola and Mieso. The inbred lines were alsornevaluated separately adjacent to the hybrid trial at each site and the inbred lines wererngenotyped by SNP and ISSR markers. Significant differences were observed among therninbred lines and hybrids for grain yield and most considered agronomic traits indicating thatrngenetic variations existed among the genotypes to allow good progress from selection forrnimprovements of those traits. Across sites, the highest mean grain yield was observed for thernhybrid L52/CML159 (5.38 t ha-1) followed by L18/CML159 (5.07 t ha-1) and L35/CML159rn(5.02 t ha-1) in the hybrid trial while inbred line L52 showed the highest mean GY whichrnwas 3.15 t ha-1 followed by L38 (2.94 t ha-1), L47 (2.88 t ha-1), L17 (2.43 t ha-1), L48 (2.33 trnha-1) and L40 (2.06 t ha-1) in the inbred line trial. The combining ability analysis showedrnthat general combining ability (GCA) of lines was significant while specific combiningrnability (SCA) was non-significant for grain yield, anthesis date, plant height, ear height,rnplant aspect, ear length and thousand kernel weight indicating that the variability observedrnamong the hybrids was attribuable to additive effects for most traits. The contribution of linernivrnGCA was found to be higher than the contribution of tester GCA and line ï‚´ tester SCA forrnall considered traits except for thousand kernel weight where the contribution of tester GCArnwas higher. Inbred lines L35 (0.83 t ha-1), L45 (0.68 t ha-1), L53 (0.63 t ha-1), L4 (0.57 trnha-1), L21 (0.56 t ha-1), L52 (0.54 t ha-1) and L32 (0.49 t ha-1) had significant positive GCArneffects for grain yield. Hybrid combination L52/CML159 had the best SCA effects for grainrnyield and other most important traits and the maximum standard heterosis over MH140rn(19.7%) and MH130 (21.4%). Grain yield had positive and highly significant genotypic andrnphenotypic correlations with plant height, ear length and thousand kernel weight whilernnegative and highly significant with days to 50% anthesis, anthesis-silking interval, earrnposition, shoot lodging and ear aspect. The present study showed that root lodging was notrnimportant characteristics to be considered while shoot lodging played an important role inrndetermining grain yield. It also showed that ear height is more important than plant height torndevelop high yielding hybrids and that it is possible to select high yielding varieties whichrnare early but tall with low ear placement. The ISSR markers were found to be as effective asrnSNP markers in clustering inbred lines into seven sub-groups which are in agreement withrnpedigree information. All the three multivariate analyses viz. cluster analysis, model-basedrnpopulation structure analysis and principal component analysis using SNPs consistentlyrnidentified the same seven distinct populations and revealed similar membership of inbredrnlines in each population. In general, the results from this diversity study based on SNP andrnISSR markers will be useful to breeders in selecting best parental combinations for startingrnnew breeding populations, mapping population and marker assisted breeding.