Background: Drug-induced liver injury (DILI) is a well-recognized adverse effect of anti-rntubercular drugs (ATD) and antiretroviral therapy (ART) possibly associated with genetic rnvariations. Liver injury induced by these drugs carries numerous risks that include morbidity and rnmortality due to liver failure, disease progression resulting from discontinuation of drug therapy, rnand drug resistance related to treatment interruption. Genome-wide association study (GWAS) is rna genetic study used to identify single nucleotide polymorphisms (SNPs) that contribute to rndisease risk. Genetic risk variants identified by pharmacogenomic GWAS, and confirmed rnthrough replication studies can serve as biomarkers to predict treatment response, and advance rnclinical care through personalized medicine. The objective of this study was to identify and rnvalidate SNPs associated with ATD and ART-induced liver injury through GWAS and rnsubsequent replication studies, and also to investigate the association of human leukocyte antigen rn(HLA)-B alleles with ATD and ART co-treatment induced liver injury in Ethiopian TB/HIV co-rninfected patients. rnMethods: The present study had a case-control design using cases and controls obtained from a rnprospective cohort study conducted to determine the incidence and predictors for ATD and ART-rninduced liver injury in Ethiopia. Cases were those study participants who developed DILI due to rnATD and/or ART in the follow up period. DNA samples from a total of 1,055 patients were rnused. The three treatment groups for the current study were TB patients treated with first-line rnATD (75 cases, 571 controls), HIV patients without TB co-infection treated with Efavirenz-rnbased ART (21 cases, 368 controls), and TB/HIV co-infected patients treated with both ATD and rnefavirenz-based ART concurrently (46 cases, 292 controls). Whole genome genotyping was done rnusing Illumina Omni Express Exome BeadChip genotyping array on a total 89 cases and 488 controls. Replication study was carried out for the top SNPs with the lowest P-values in the rnGWAS using an independent cohort consisting of 45 cases and 425 controls. For the HLA study, rngenomic DNA from 46 cases, and sex and age matched controls from ATD and ART co-rntreatment group were typed for HLA-B alleles using low resolution Olerup SSPrntyping kit. High resolution sub-typing was performed for HLA-B alleles that showed significant rnassociation with DILI. The association analysis for the GWAS and replication studies were done rnusing Plink v 1.07, and we used SPSS v 22 for the HLA-B typing analysis. rnResults: In the combined analysis of the GWAS and the replication study, the top SNP identified rnin the ATD treatment group was rs10946737 (P = 4.4Ã—10rn-6rnÂ®rnHLA-B DNA rn, odds ratio (OR) = 3.4, 95% CI = 2.2-rn5.3) located in the intron region of family with sequence similarity-65 member-B (FAM65B). In rnaddition, we identified cluster of SNPs with suggestive genome-wide significance in the intron of rnATP/GTP binding protein like-4 (AGBL4). We identified a missense SNP rs199650082 (R919Q, rnP = 1.4Ã—10rn-6rn, OR = 18.2, 95% CI = 7.1-46.9) in an endoplasmic reticulum to nucleus signaling-1 rn(ERN1) gene in the ART group. In the ATD and ART co-treatment group, we identified rnrs4842407 (P = 5.3Ã—10rn-7rn, OR = 5.4, 95% CI = 2.8-10.3) a long intergenic non-coding RNA rn(lincRNA) transcript variant. In our HLA typing study, proportion of HLA-Brnthe cases was significantly higher compared with the controls (P = 0.002). From HLA-Brnalleles detected, HLA-Brn*rn57:02 and HLA-Brn*rn*rn57 allele carriers in rn57:03 accounted for 41.7 and 58.3%, respectively. rnConclusion: We identified genetic variants that may be associated with ATD and ART-induced rnliver injuries. Further studies with larger sample sizes are essential to confirm the findings.