Genetic variants of CYP2C19 and CYP2C18 and their clinical implications in the Saudi population.

The authors extend their appreciation to the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia (Grant Number: KFU261172).

All the authors made substantial contributions to the conception of the study. AK: conceptualization, methodology, investigation, data curation, formal analysis, writing—original draft; AH and YM: investigation, resources, formal analysis, writing—original draft; MK: resources, validation, writing—review and editing, supervision; AT: conceptualization, methodology, data curation, formal analysis, writing—review and editing, supervision. All authors have critically reviewed and approved the final draft and are responsible for the content and similarity index of the manuscript.

The research protocol for this study was approved by the Research Ethics Committee of King Fahad Hospital (IRB log number 27/A-29-2020) and Research Ethics Committee of King Faisal University (IRB log number KFU-REC/2021-03-09). Before participants’ involvement in the study, the researchers obtained written informed consent. Participants' confidentiality and anonymity were ensured by assigning each participant a unique code number solely for analysis purposes. Participation in the present study was entirely voluntary.

A total of 374 blood specimens were collected across various locations in the Alahsa area. The mean participant age was 40.1 ± 13.8 years, and approximately 53 % of participants were female ( n = 199). The DNA samples were successfully isolated and met the quality assurance standards. Table S1 shows 146 genetic markers of CYP2C19 and CYP2C18 genes, and their chromosomal positions, GenTrain scores, call frequencies, numbers of calls, and numbers of no calls. We were able to measure 141 CYP2C19 and CYP2C18 variants. Most variants had high GenTrain scores (>0.8) and insignificant ChiTest100 p -values. In addition, the sample call rate was very high: more than 98 % of samples had a score of 1. Table 1 lists the 53 identified CYP2C19 and CYP2C18 variants among the 141 tested variants, along with their star alleles and genotypes (homozygous dominant, homozygous recessive, or heterozygous). Some of these variants with high MAF (>20 %), including rs7916649, rs4917623, rs2901783, rs7915414, rs11528090, rs4494250, rs2860840, rs12248560, and rs4388808, have been associated with alterations in the pharmacokinetics and pharmacodynamics of commonly prescribed drugs, such as omeprazole, clopidogrel, nevirapine, tamoxifen, voriconazole, warfarin, and valproic acid. Additionally, five variants (rs4917623, rs4494250, rs12248560, rs4388808, and rs11592737) have been associated with various health disorders, acting either as risk factors or protective factors. These disorders include breast cancer, chronic obstructive pulmonary disease, endometriosis, endometrial thickness, tetralogy of Fallot, major depressive disorder, hypertension, type 2 diabetes mellitus, Alzheimer's disease, and female infertility ( Table 2 and Table S2 ). Moreover, Table 3 shows the predictive functional consequences of the detected variants, according to computational tools. According to their DANN and SIFT scores, six (n = 6) variants, rs1126545, rs17879685, rs17885179, rs59636573, rs60181876, and rs770829708, were associated with major deterioration in CYP2C19 and CYP2C18 enzymatic activity and possibly with health issues. However, the other measured variants did not show significant in silico enzymatic deterioration. Table 1 MAF and genotype frequencies of CYP2C19 and CYP2C18 gene variants detected among Saudi participants (n = 53). Table 1 Participant SNP Star allele, nitrogenous base change MAF Homozygous genotype for minor allele, N (%) Heterozygous genotype for minor allele, N (%) Homozygous genotype for wild-type allele, N (%) 1. rs12769205 CYP2C19 ∗2, A > G 14.17 13 (3.48 %) 80 (21.39 %) 281 (75.13 %) 2. rs12571421 CYP2C19 ∗2, A > G 11.36 8 (2.14 %) 69 (18.45 %) 297 (79.41 %) 3. rs145119820 CYP2C19 ∗2, G > A 0.80 0 (0 %) 6 (1.6 %) 368 (98.4 %) 4. rs7916649 CYP2C19 ∗2B, G > A 46.66 77 (20.59 %) 195 (52.14 %) 102 (27.27 %) 5. rs770829708 CYP2C19 ∗2F, G > A 0.13 0 (0 %) 1 (0.27 %) 372 (99.73 %) 6. rs17878459 CYP2C19 ∗3, G > A 0.67 0 (0 %) 5 (1.34 %) 369 (98.66 %) 7. rs17879685 CYP2C19 ∗13, C > T 0.13 0 (0 %) 1 (0.27 %) 373 (99.73 %) 8. rs3814637 CYP2C19 ∗15, C > T 7.91 2 (0.54 %) 55 (14.75 %) 316 (84.72 %) 9. rs12248560 CYP2C19 ∗17, C > T 20.99 11 (2.94) 135 (36.1 %) 228 (61 %) 10. rs11188072 CYP2C19 ∗17, C > T 20.99 11 (2.94 %) 135 (36.1 %) 228 (60.96 %) 11. rs11188059 CYP2C18 G > A 4.55 0 (0 %) 34 (9.09 %) 340 (90.91 %) 12. rs7902257 CYP2C19 ∗27, G > A 2.28 1 (0.27 %) 15 (4.01 %) 358 (95.72 %) 13. rs111490789 CYP2C19 ∗28, C > A 0.41 0 (0 %) 3 (0.81 %) 367 (99.19 %) 14. rs4417205 CYP2C19 ∗35, C > G 14.17 13 (3.48 %) 80 (21.39 %) 281 (75.13 %) 15. rs17885179 CYP2C19 ∗39, A > C 0.40 0 (0 %) 3 (0.8 %) 371 (99.2 %) 16. rs28399513 CYP2C19 , T > A 12.17 10 (2.67 %) 71 (18.98 %) 293 (78.34 %) 17. rs11528090 CYP2C19 , T > G 21.43 17 (4.58 %) 125 (33.69 %) 228 (61.73 %) 18. rs12768009 CYP2C19 , G > A 12.17 10 (2.67 %) 71 (18.98 %) 293 (78.34 %) 19. rs2860840 CYP2C18 , C > T 21.12 16 (4.27 %) 126 (33.68 %) 232 (62.03 %) 20. rs1126545 CYP2C18 , C > T 11.36 8 (2.1 %) 69 1 (8.45 %) 297 (79.4 %) 21. rs4986894 CYP2C19 ∗2C, T > C 11.36 8 (2.14 %) 69 (18.45 %) 297 (79.41 %) 22. rs12268020 CYP2C19 , C > T 20.99 11 (2.94 %) 135 (36.1 %) 228 (60.96 %) 23. rs4917623 CYP2C19 , T > C 42.78 68 (18.18 %) 184 (49.2 %) 122 (32.62 %) 24. rs11592737 CYP2C19 , A > G 20.99 11 (2.94 %) 135 (36.1 %) 228 (60.96 %) 25. rs7067866 CYP2C19 , G > T 46.79 77 (22.59 %) 196 (52.41 %) 101 (27.01 %) 26. rs7917985 CYP2C18 , C > A 23.53 16 (4.28 %) 144 (38.5 %) 214 (57.22 %) 27. rs2901783 CYP2C18 , A > G 22.06 22 (5.88 %) 121 (32.35 %) 231 (61.76 %) 28. rs4494250 CYP2C19 , G > A 21.26 17 (4.55 %) 125 (33.42 %) 232 (62.03 %) 29. rs4388808 CYP2C19 , A > G 20.99 16 (4.28 %) 125 (33.42 %) 233 (62.3 %) 30. rs11188092 CYP2C19 , A > C 20.72 10 (2.67 %) 135 (36.1 %) 229 (61.23 %) 31. rs7085563 CYP2C18 , T > A 16.58 15 (4.01 %) 94 (25.13 %) 265 (70.86 %) 32. rs2260946 CYP2C19 , T > C 14.71 13 (3.48 %) 84 (22.46 %) 277 (74.06 %) 33. rs7900135 CYP2C18 , A > G 14.04 13 (3.48 %) 79 (21.12 %) 282 (75.4 %) 34. rs932809 CYP2C18 , C > T 14.04 13 (3.48 %) 79 (21.12 %) 282 (75.4 %) 35. rs3740367 CYP2C18 , G > A 4.81 1 (0.27 %) 34 (9.09 %) 339 (90.64 %) 36. rs117140172 CYP2C18 , G > A 2.01 1 (0.27 %) 13 (3.48 %) 360 (96.26 %) 37. rs1326830 CYP2C18 , C > A 2.01 0 (0 %) 15 (4.01 %) 359 (95.99 %) 38. rs17884832 CYP2C19 , T > G 6.95 2 (0.53 %) 48 (12.83 %) 324 (86.63 %) 39. rs17879992 CYP2C19 , T > C 7.37 2 (0.54 %) 51 (13.67 %) 321 (85.79 %) 40. rs187091323 CYP2C19 , A > G 4.58 1 (0.27 %) 32 (8.63 %) 341 (91.11 %) 41. rs4917612 CYP2C19 , C > G 9.09 6 (1.6 %) 56 (14.97 %) 312 (83.42 %) 42. rs733115 CYP2C19 , G > T 6.95 2 (0.53 %) 48 (12.83 %) 324 (86.63 %) 43. rs941890 CYP2C19 , G > A 2.54 1 (0.27 %) 17 (4.55 %) 356 (95.19 %) 44. rs150579865 CYP2C18 , G > T 0.53 0 (0 %) 4 (1.07 %) 370 (98.93 %) 45. rs60181876 CYP2C18 , C > T 0.67 0 (0 %) 5 (1.34 %) 369 (98.66 %) 46. rs59636573 CYP2C18 , G > T 0.13 0 (0 %) 1 (0.27 %) 373 (99.73 %) 47. rs113164681 CYP2C19 , C > T 0.94 0 (0 %) 7 (1.87 %) 367 (98.13 %) 48. rs17878739 CYP2C19 , T > C 0.94 0 (0 %) 7 (1.87 %) 367 (98.13 %) 49. rs17878649 CYP2C19 , G > A 0.27 0 (0 %) 2 (0.53 %) 372 (99.47 %) 50. rs141417293 CYP2C19 , C > A 0.27 0 (0 %) 2 (0.53 %) 372 (99.47 %) 51. rs143612134 CYP2C19 , T > C 0.27 0 (0 %) 2 (0.53 %) 372 (99.47 %) 52. rs150790215 CYP2C19 , G > A 0.80 0 (0 %) 6 (1.6 %) 368 (98.4 %) 53. rs7915414 CYP2C19 , G > A 22.01 22 (5.98 %) 120 (32.07 %) 232 (61.96 %) Table 2 Effects of the most frequently detected (MAF >20 %) CYP2C19 and CYP2C18 gene polymorphisms on human health and the metabolism of several drugs. Table 2 Participant Variant Gene MAF Predicted enzymatic activity Drugs Effects of SNPs on drugs Clinical implications 1. rs7916649 CYP2C19 46.66 Inactive Valproic acid and clopidogrel Effects on valproic acid: increased risk of valproic acid toxicity. NA Effects on clopidogrel: Increased risk of recurrent cardiac events due to treatment failure 2. rs4917623 CYP2C19 42.78 Decreased Tamoxifen and nevirapine Effects on tamoxifen: Enhanced anti-cancer efficacy. The CYP2C19 rs4917623 heterozygous genotype is more strongly associated with breast cancer (odds ratio = 1.38) than the wild-type homozygous genotype. Effects on nevirapine: rs4917623 allele might be associated with elevated trough plasma concentrations of nevirapine, and might lead to enhanced efficacy as well as adverse effects. 3. rs2901783 CYP2C18 22.06 Unconfirmed Clopidogrel and warfarin Effects on clopidogrel: rs2901783 variant is significantly associated with a lower clopidogrel efficacy than the referenced allele. NA Effects on warfarin: rs2901783 is associated with accelerated achievement of a therapeutic INR. 4. rs7915414 CYP2C19 22.01 Unconfirmed Clopidogrel Effects on clopidogrel: rs7915414 variant is significantly associated with lower clopidogrel efficacy than the referenced allele. NA 5. rs11528090 CYP2C19 21.43 Unconfirmed; Potentially associated with decreased enzymatic activity Nevirapine Effects on nevirapine: rs11528090 allele might be associated with elevated trough plasma concentrations of nevirapine, and might lead to enhanced efficacy as well as adverse effects. NA 6. rs4494250 CYP2C19 21.26 Unconfirmed; Potentially associated with decreased enzymatic activity Anti-HIV drugs (NNRTIs and PIs) Effects on anti-HIV drugs (NNRTIs and PIs): carriers of the rs4494250 allele might have enhanced anti-HIV treatment efficacy by reducing viral load, if treated with NNRTI and PI drugs. The rs4494250 allele might be associated with greater breast cancer risk than the wild-type allele. The rs4494250 allele is associated with enhanced risk of elevated diastolic blood pressure. The rs4494250 minor allele might be associated with greater chronic obstructive pulmonary disease risk than the wild-type allele. 7. rs2860840 CYP2C18 21.12 Unconfirmed; potentially associated with increased enzymatic activity Omeprazole, voriconazole, and warfarin Effects on omeprazole: rs2860840 allele is associated with treatment failure and might necessitate increased doses. NA Effects on voriconazole: rs2860840 allele is associated with treatment failure and might necessitate increased doses. Effects on warfarin: rs2860840 allele is associated with diminished warfarin efficacy due to rapid metabolism. 8. rs12248560 CYP2C19 20.99 Increased Clopidogrel, prasugrel, and tamoxifen Effects on clopidogrel and prasugrel: carriers of the rs12248560 T allele have a significantly higher rate of hyper-response to clopidogrel and higher rates of bleeding complications than carriers of the CC genotype. Effects on tamoxifen: Women with breast cancer carrying the rs12248560 variant might have diminished response to tamoxifen. Women carrying the rs12248560 T allele (CT and TT genotypes) have diminished risk of endometriosis and breast cancer, potentially in association with enhanced estradiol catabolism. Pregnant individuals carrying the rs12248560 T allele (CT and TT genotypes) might have elevated risk of delivering newborns with tetralogy of Fallot. Carriers of the rs12248560 T allele have elevated risk of hypertension and type-2 diabetes. 9. rs11592737 CYP2C19 20.99 Unconfirmed; potentially associated with normal enzymatic activity NA NA Women carrying the rs11592737 GG genotype have greater risk of endometriosis and infertility than those carrying the rs11592737 A allele. Women with the rs11592737 GG genotype have greater risk of infertility and endometriosis than those carrying the rs11592737 A allele. 10. rs4388808 CYP2C19 20.99 Unconfirmed; potentially associated with decreased enzymatic activity Nevirapine Effects on nevirapine: rs4388808 G allele might be associated with elevated trough plasma concentrations of nevirapine and might lead to enhanced efficacy as well as adverse effects. Carriers of the rs4388808 G allele have a significantly lower amyloid-β load in the frontal, inferior temporal, and posterior cingulate cortices in the brain than AA genotype carriers. Therefore, the rs4388808 G allele is recognized as a protective genetic factor against Alzheimer's disease. NA: No relevant data or information is available. NNRTIs and PIs: Non-nucleoside reverse transcriptase inhibitors and protease inhibitors. Table 3 Functional consequences of the detected variants (n = 53). Table 3 Participant Variant ID Gene Sequence ontology Protein sequence change SIFT score a CADD score b DANN score c regSNP-intron Ω VEST score ¶ Predicted enzymatic activity 1. rs111490789 CYP2C19 Non-coding – – 0.681 0.2773 – – Unconfirmed 2. rs11188059 CYP2C18 Non-coding – – 7.594 0.5633 0.5 – Increased 3. rs11188072 CYP2C19 Non-coding – – 0.25 0.3745 – – Increased 4. rs11188092 CYP2C19 Non-coding – – 1.331 0.3358 0.67 – Unconfirmed 5. rs1126545 CYP2C18 Missense T385M 0.034 15.28 0.9723 – 0.318 Inactive 6. rs113164681 CYP2C19 Non-coding – – 0.479 0.383 – – Unconfirmed 7. rs11528090 CYP2C19 Non-coding – – 1.56 0.3714 0.14 – Unconfirmed; Potentially associated with decreased enzymatic activity 8. rs11592737 CYP2C19 Non-coding – – 0.313 0.7489 0.63 – Unconfirmed; potentially associated with normal enzymatic activity 9. rs117140172 CYP2C18 Non-coding – – 2.138 0.4199 0.25 – Unconfirmed 10. rs12248560 CYP2C19 Regulatory – – 0.233 0.3703 – – Increased 11. rs12268020 CYP2C19 Non-coding – – 0.269 0.3836 0.67 – Unconfirmed; potentially associated with normal enzymatic activity 12. rs12571421 CYP2C19 Non-coding – – 3.321 0.7417 0.55 – Inactive 13. rs12768009 CYP2C19 Non-coding – – 1.909 0.7394 0.46 – Unconfirmed; Potentially associated with decreased enzymatic activity 14. rs12769205 CYP2C19 Non-coding – – 1.03 0.6396 0.26 – Inactive 15. rs1326830 CYP2C18 Non-coding – – 0.193 0.2623 – – Unconfirmed 16. rs141417293 CYP2C19 Non-coding – – 2.043 0.1969 0.45 – Unconfirmed 17. rs143612134 CYP2C19 Non-coding – – 3.585 0.6413 0.77 – Unconfirmed 18. rs145119820 CYP2C19 Missense V113I 1 0.002 0.8678 – 0.176 Decreased 19. rs150579865 CYP2C18 Non-coding – – 0.46 0.2657 0.66 – Unconfirmed 20. rs150790215 CYP2C19 Non-coding – – 3.571 0.3808 0.52 – Unconfirmed 21. rs17878459 CYP2C19 Missense E92D 0.365 1.085 0.6485 – – Inactive 22. rs17878649 CYP2C19 Non-coding – – 2.796 0.2572 0.73 – Unconfirmed 23. rs17878739 CYP2C19 Non-coding – – 0.173 0.2972 – – Unconfirmed 24. rs17879685 CYP2C19 Missense R410C 0.037 16.42 0.08931 – 0.264 Unconfirmed 25. rs17879992 CYP2C19 Non-coding – – 1.02 0.6592 0.52 – Unconfirmed 26. rs17884832 CYP2C19 Non-coding – – 5.644 0.7492 0.39 – Unconfirmed 27. rs17885179 CYP2C19 Missense E122A 0.281 19.47 0.9636 - 0.52 Unconfirmed; Potentially associated with decreased enzymatic activity 28. rs187091323 CYP2C19 Non-coding – – 1.662 0.09841 0.48 – Unconfirmed 29. rs2260946 CYP2C19 Non-coding – – 1.639 0.3275 0.24 – Unconfirmed 30. rs28399513 CYP2C19 Non-coding – – 0.418 0.2191 0.25 – Unconfirmed 31. rs2860840 CYP2C18 Non-coding – – 1.686 0.5899 – – Unconfirmed; Potentially associated with increased enzymatic activity 32. rs2901783 CYP2C18 Non-coding – – 0.182 0.4646 0.32 – Unconfirmed 33. rs3740367 CYP2C18 Non-coding – – 0.204 0.2274 – – Unconfirmed 34. rs3814637 CYP2C19 Non-coding – – 4.773 0.5647 – – Inactive 35. rs4388808 CYP2C19 Non-coding – – 4.562 0.8193 0.26 – Unconfirmed; Potentially associated with decreased enzymatic activity 36. rs4417205 CYP2C19 Non-coding – – 5.274 0.6617 0.21 – Unconfirmed 37. rs4494250 CYP2C19 Non-coding – – 1.011 0.3958 0.32 – Unconfirmed; Potentially associated with decreased enzymatic activity 38. rs4917612 CYP2C19 Non-coding – – 0.054 0.3323 0.09 – Unconfirmed 39. rs4917623 CYP2C19 Non-coding – – 0.243 0.3293 0.97 – Decreased 40. rs4986894 CYP2C19 Non-coding – – 2.632 0.5392 – – Unconfirmed 41. rs59636573 CYP2C18 Missense V330L 0.001 18.8 0.9871 – 0.666 Unconfirmed 42. rs60181876 CYP2C18 Missense T299I 0.025 23.6 0.9979 – 0.453 Unconfirmed 43. rs7067866 CYP2C19 Non-coding – – 0.628 0.1727 – – Unconfirmed; Potentially associated with increased enzymatic activity 44. rs7085563 CYP2C18 Non-coding – – 1.291 0.2497 0.38 – Unconfirmed 45. rs733115 CYP2C19 N/K – – 3.121 0.4313 – – Unconfirmed 46. rs770829708 CYP2C19 Missense D341N 0 26 0.9992 – 0.39 Decreased 47. rs7900135 CYP2C18 Non-coding – – 8.62 0.8819 0.23 – Unconfirmed 48. rs7902257 CYP2C19 Regulatory – – 0.186 0.7151 – – Decreased 49. rs7915414 CYP2C19 Non-coding – – 0.408 0.3888 0.79 – Unconfirmed 50. rs7916649 CYP2C19 Non-coding – – 1.204 0.2696 0.28 – Inactive 51. rs7917985 CYP2C18 Non-coding – – 0.432 0.6275 0.51 – Unconfirmed; potentially associated with normal enzymatic activity 52. rs932809 CYP2C18 Non-coding – – 2.372 0.3141 0.67 – Unconfirmed 53. rs941890 CYP2C19 N/K – – 0.014 0.1292 – – Unconfirmed Ω regSNP-intron: This score is only for intronic variants. Scores <0.05 indicate a damaging protein, whereas scores between 0.05 and 0.1 indicate a possibly damaging protein. ¶ VEST (variant effect scoring tool) score: This score is only for exonic variants. Scores range from 0.0 to 1.0. A score >0.5 indicates a possible deleterious variant. N/K: Unknown. a SIFT (sorting intolerant from tolerant) score: This score is only for exonic variants. Scores range from 0.0 to 1.0. A score ≤0.05 indicates a possible deleterious variant. b CADD (combined annotation-dependent depletion) score: Scores ≥10 indicate that the variant is among the 10 % most deleterious substitutions affecting the human genome. c DANN (deleterious annotation of genetic variants with neural network) score: Scores range from 0 to 1. A score >0.5 indicates a possible deleterious variant. MAF and genotype frequencies of CYP2C19 and CYP2C18 gene variants detected among Saudi participants (n = 53). Effects of the most frequently detected (MAF >20 %) CYP2C19 and CYP2C18 gene polymorphisms on human health and the metabolism of several drugs. NA: No relevant data or information is available. NNRTIs and PIs: Non-nucleoside reverse transcriptase inhibitors and protease inhibitors. Functional consequences of the detected variants (n = 53). Ω regSNP-intron: This score is only for intronic variants. Scores <0.05 indicate a damaging protein, whereas scores between 0.05 and 0.1 indicate a possibly damaging protein. ¶ VEST (variant effect scoring tool) score: This score is only for exonic variants. Scores range from 0.0 to 1.0. A score >0.5 indicates a possible deleterious variant. N/K: Unknown. SIFT (sorting intolerant from tolerant) score: This score is only for exonic variants. Scores range from 0.0 to 1.0. A score ≤0.05 indicates a possible deleterious variant. CADD (combined annotation-dependent depletion) score: Scores ≥10 indicate that the variant is among the 10 % most deleterious substitutions affecting the human genome. DANN (deleterious annotation of genetic variants with neural network) score: Scores range from 0 to 1. A score >0.5 indicates a possible deleterious variant. Table 4 presents the LD among the 53 observed genetic variants according to their D′, r 2 , and LOD values. The variants rs7900135, rs932809, rs7085563 ( CYP2C18 variants), and rs12768009 (a CYP2C19 variant) were strongly associated with, and inherited together with, the rs12769205 ( CYP2C19 ∗2E) variant related to CYP2C19 enzyme inactivity. Furthermore, the rs7917985 (a CYP2C18 variant), rs11188072, rs11188092, rs11592737, and rs12268020 variants correlated with the rs12248560 variant (also called CYP2C19 ∗17 and associated with CYP2C19 ultra-activity). In addition, the rs28399513 and rs12768009 variants were associated with the rs12571421 variant (also called CYP2C19 ∗2 and associated with inactive CYP2C19). Moreover, rs7067866, a highly detected CYP2C19 variant (MAF = 46.8 %), was strongly associated with rs7916649 ( CYP2C19 ∗2B). Table 4 D′, logarithm of the odds, correlation coefficient, and 95 % confidence intervals for SNP pairs. Table 4 Participant SNP-1 SNP-2 D′ LOD r 2 95 %CI lower 95 %CI upper 1. rs7900135 rs12769205 1 105.53 0.989 0.97 1 2. rs7917985 rs12248560 1 101.62 0.863 0.97 1 3. rs932809 rs12769205 1 105.53 0.989 0.97 1 4. rs7085563 rs12769205 0.988 83.1 0.812 0.94 1 5. rs1126545 rs12768009 1 85.24 0.926 0.96 1 6. rs11188072 rs12248560 1 126.27 1 0.98 1 7. rs113164681 rs111490789 1 7.57 1 0.6 1 8. rs111490789 rs17878739 1 7.57 1 0.6 1 9. rs3814637 rs17884832 1 58.15 0.872 0.94 1 10. rs4986894 rs12768009 1 85.24 0.926 0.96 1 11. rs7067866 rs7916649 1 162.46 0.995 0.98 1 12. rs17884832 rs17879992 1 62.48 0.941 0.95 1 13. rs4388808 rs11528090 1 123.11 0.976 0.98 1 14. rs12571421 rs28399513 1 85.24 0.926 0.96 1 15. rs12768009 rs12571421 1 85.24 0.926 0.96 1 16. rs2860840 rs4494250 1 127.22 0.992 0.98 1 17. rs4417205 rs28399513 1 80.73 0.839 0.96 1 18. rs3740367 rs187091323 1 41.93 0.964 0.92 1 19. rs11188092 rs11592737 1 120.76 0.984 0.98 1 20. rs11188092 rs12268020 1 120.76 0.984 0.98 1 21. rs12248560 rs11188092 1 120.76 0.984 0.98 1 22. rs932809 rs28399513 0.987 77.86 0.826 0.94 1 23. rs12768009 rs12769205 1 80.73 0.839 0.96 1 24. rs12768009 rs28399513 1 98.89 1 0.98 1 25. rs11592737 rs12268020 1 126.27 1 0.98 1 26. rs7917985 rs11592737 1 101.62 0.863 0.97 1 27. rs11188072 rs11592737 1 126.27 1 0.98 1 28. rs12248560 rs11592737 1 126.27 1 0.98 1 29. rs59636573 rs17879685 1 3.01 1 0.28 1 30. rs11188072 rs12268020 1 126.27 1 0.98 1 31. rs12248560 rs12268020 1 126.27 1 0.98 1 32. rs17884832 rs733115 1 67.54 1 0.96 1 33. rs7902257 rs941890 1 27.95 0.892 0.87 1 D′ : value of D′ between loci. LOD : log of the likelihood odds ratio, a measure of confidence in the value of D′. r 2 : correlation coefficient between loci. 95 %CI lower : 95 % confidence interval of the lower bound on D′. 95 %CI upper : 95 % confidence interval of the upper bound on D′. D′, logarithm of the odds, correlation coefficient, and 95 % confidence intervals for SNP pairs. D′ : value of D′ between loci. LOD : log of the likelihood odds ratio, a measure of confidence in the value of D′. r 2 : correlation coefficient between loci. 95 %CI lower : 95 % confidence interval of the lower bound on D′. 95 %CI upper : 95 % confidence interval of the upper bound on D′. We additionally identified six LD blocks with D′ ≥ 8 ( Figure 1 , Figure 2 ). The first was the largest block and the only block involving CYP2C18 variants. It consisted of eight SNPs: six CYP2C18 variants with inconclusive effects and two CYP2C19 variants associated with ultra-metabolism. The strong linkage suggested that these specific CYP2C18 variants were likely co-segregate with CYP2C19 variants associated with elevated enzymatic activity. In this LD block, three haplotypes (CTCCGCGC, CTCTGCGC, and CTCCGCTT) were present among >20 % of the included participants ( Figure 3 ). Figure 1 The first three LD blocks for CYP2C19 and CYP2C18 SNPs detected in our study. Figure 1 Figure 2 The second three LD blocks for CYP2C19 and CYP2C18 SNPs detected in our study. Figure 2 Figure 3 Haplotype structure across the six determined LD blocks. Figure 3 The first three LD blocks for CYP2C19 and CYP2C18 SNPs detected in our study. The second three LD blocks for CYP2C19 and CYP2C18 SNPs detected in our study. Haplotype structure across the six determined LD blocks. LD block 2 was dominated by the CCTC haplotype, which involved rs3814637, a variant associated with a poor metabolizer phenotype. LD block 3 included variants with mixed effects on the CYP2C19 enzyme. The most frequently observed haplotypes (CTG and TTG) were associated with rapid-metabolizer and ultra-metabolizer phenotypes. Although less commonly detected than the two aforementioned haplotypes in LD block 3, the CCA haplotype, associated with the poor metabolizer phenotype, was observed among 11 % of participants ( Figure 1 , Figure 3 ). The remaining LD blocks (blocks 4–6) consisted predominantly of variants associated with the inactive CYP2C19 enzyme. The minor allele frequencies among the identified haplotypes ranged between 11 % and 13 % ( Figure 2 , Figure 3 ).

This study investigated the presence of many genetic variants associated with the CYP2C19 and CYP2C18 genes that were previously unreported in the Middle Eastern population, including the Saudi population. The most common (MAF >10 %) and important detected genetic polymorphisms were rs12571421, rs12248560, rs7916649, rs12769205, rs11188072, rs2860840, and rs4917623, which are likely to affect the metabolism of essential drugs for treating severe cases of epilepsy, fungal infections, cardiovascular diseases, kidney failure, and psychiatric diseases. Nonetheless, further pharmacogenomics investigations will be necessary to validate the effects of these variants and haplotypes through high-quality in vitro, in vivo, and clinical association studies. The current clinical guidelines for pharmacogenomics and medical genetics, such as those from the Clinical Pharmacogenetics Implementation Consortium and the American College of Medical Genetics and Genomics, have relied on studies conducted among non-Middle Eastern populations. Consequently, several recommendations regarding CYP2C19 and CYP2C18 genetic variants might not precisely reflect the Middle Eastern genetic profile. Our findings establish an important local baseline and suggest that locally adapted guidelines will be essential to focus on the observed haplotypes in future research. By integrating the current outcomes associated with variants with established associations, the healthcare systems of KSA and other Middle Eastern countries could optimize prophylactic measures, drug selection, and dosing for CYP2C19 and CYP2C18 substrates, thereby moving toward effective application of personalized medicine.

This study examined 141 SNPs associated with the CYP2C19 and CYP2C18 genes and their allele and genotype frequencies in 374 Saudi individuals. The MAFs ranged from 0.13 % to 46.79 % for 53 SNPs. Several SNPs were not detected (n = 88) or had call rates <95 % (n = 5). Thirty-one (n = 31) SNPs were very common (MAF ≥5 %), including rs7067866 (47 %), rs7916649 (47 %), rs4917623 (43 %), rs7917985 (24 %), rs7915414 (22 %), rs11528090 (21 %), rs4494250 (21 %), rs2860840 (21 %), rs2901783 (22 %), and rs12268020 (22.0 %). The remaining 22 SNPs (MAF <5 %) were less common, including rs117140172 (2 %), rs1326830 (2 %), rs7902257 (2 %), rs941890 (3 %), rs11188059 (4 %), rs113164681 (1 %), rs150790215 (1 %), rs145119820 (1 %), rs60181876 (1 %), and rs17878459 (1 %). Haplotype analysis identified two CYP2C19 variants with MAF >10 % (rs4417205 and rs12768009) that were significantly associated with other well-known variants in a single haplotype. Additionally, four CYP2C18 variants (rs932809, rs7085563, rs1126545, and rs1326830) were associated with one CYP2C19 variant (rs7067866) previously suggested to be associated with the ultra-rapid metabolizer phenotype. 19 Furthermore, the dominant haplotypes were associated with rapid-metabolizer phenotypes, followed by ultra-rapid metabolizer and poor-metabolizer phenotypes. Of note, in prior studies, CYP2C18 protein expression has been poorly detected in liver tissue but detected in other tissues, such as skin and lung, thus suggesting potential extrahepatic involvement of CYP2C18 in the metabolism of certain drugs. Moreover, some studies have associated certain CYP2C18 variants with the metabolism of clopidogrel, warfarin, and phenytoin 20 , 21∗ , 22 . Several studies in the Middle East have examined the occurrence of common CYP2C19 variants, including rs4244285 ( CYP2C19∗2 ), rs4986893 ( CYP2C19∗3 ), rs17878459 ( CYP2C19∗3 ), rs28399504 ( CYP2C19∗4 ), rs56337013 ( CYP2C19∗5 ), rs2558184 ( CYP2C19∗6 ), rs72558186 ( CYP2C19∗7 ), rs41291556 ( CYP2C19∗8 ), rs17884712 ( CYP2C19∗9 ), rs6413438 ( CYP2C19∗10 ), rs55640102 ( CYP2C19∗12 ), rs17879685 ( CYP2C19∗13 ), rs55752064 ( CYP2C19∗14 ), rs17882687 ( CYP2C19∗15 ), rs12248560 ( CYP2C19∗17 ), and rs3758581 ( CYP2C19 ∗1B). However, to our knowledge, no other gene variants associated with CYP2C19 and its cognate CYP2C18 have been assessed in Middle Eastern nations. Therefore, this study is the first, to our knowledge, to determine the presence of 141 genetic variants associated with the CYP2C19 and CYP2C18 genes in the Middle Eastern population. According to the Single Nucleotide Polymorphism Database (dbSNP), several detected variants were associated with other populations ( Table S3 ). 23 For example, the MAF of rs7916649 ( CYP2C19 ∗2B), a CYP2C19 mutation associated with CYP2C19 loss-of-function, was similar to that in the global, European, and East Asian populations, but was less than that in African and Southeast Asian populations, and far higher than that in Latin American populations. This mutation was not measured in any prior study conducted in the Middle East. In addition, in the Saudi population, the MAFs of rs12248560 and rs11188072, which are CYP2C19 mutations associated with enhanced CYP2C19 function, were very close to the reference data for global, European, and African populations, but far exceeded those in East Asian, South Asian, and Latin American populations. Furthermore, we detected the rs17878459 ( CYP2C19 ∗3) variant among 0.7 % of study participants. This variant had not been detected or measured in previous studies in other Middle Eastern populations. 6 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 Additionally, the variant rs12248560 ( CYP2C19 ∗17) had a frequency of 21 %, a value lower than those in other Saudi studies but comparable to findings from Tunisian and Iraqi studies. Table S4 shows the CYP2C19 star allele frequencies in our study compared with previous studies in Middle Eastern populations, including inhabitants of the Gulf region, Iraqis, Iranians, Israelis, Jordanians, Palestinians, Egyptians, Tunisians, and Lebanese. 6 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 The previously poorly studied rs7916649 CYP2C19 variant was a highly detected CYP2C19 variant among the Saudi population. The rs7916649 variant is strongly associated with the CYP2C19 ∗2B and CYP2C19 ∗2C alleles and is probably associated with poor CYP2C19 enzymatic activity. 33 , 34 According to Malki et al., the phenotype of needing valproic acid dose reduction is common among carriers of the rs7916649 variant (composite odds ratio of 1.95 ( p -value <0.001) in carriers). 33 , 34 Because Malki et al. used large samples from British biobanks linked to medical data, the study has provided relatively strong evidence that the rs7916649 variant is associated with poor valproic acid metabolism, thereby increasing exposure to adverse drug reactions. 33 , 34 Moreover, in a 2015 study, 275 Chinese patients who carried the rs7916649 variant and used long-term clopidogrel had significantly greater risk of recurrent cardiac events, possibly because of treatment failure, than those with the homozygous CYP2C19 wild-type genotype ( p -value = 0.013). Despite the small sample size in that study, several studies have also suggested the rs7916649 variant as a possible reason for the poor metabolism of drugs reacting with the CYP2C19 enzyme. 35 Patients diagnosed with coronary artery disease and neurological disorders frequently receive prescriptions for clopidogrel and valproic acid in KSA. Consequently, the likelihood of treatment failure and adverse outcomes might be predicted according to the identification of the rs7916649 mutation in >46 % of our study sample. Nonetheless, local pharmacogenomics investigations are necessary to Several studies have suggested that the rs4917623 variant is associated with poor CYP2C19 enzymatic activity. Yoshimoto et al. 36 conducted a study that included 329 patients diagnosed with breast cancer who received tamoxifen treatment. The study revealed that patients carrying the rs4917623 TC and CC genotypes of the CYP2C19 gene had enhanced anti-cancer efficacy for longer periods than the TT genotype. However, the difference was insignificant ( p -value = 0.064). In addition, in a Chinese study of 105 patients who underwent kidney transplants and received tacrolimus as an immunosuppressant, the drug pharmacokinetics significantly differed between carriers of the rs4917623 variant and CYP2C19 wild-type allele carriers. 37 The rs4917623 variant was identified in 42.78 % of the Saudi population. Therefore, if local clinical trials confirm this association, Saudi individuals carrying the rs7916649 variant might require tacrolimus monitoring. The rs2860840 variant is known to be associated with ultra-rapid metabolism of drugs biotransformed by CYP2C19 and CYP2C18 enzymes. 38 , 39 Kee et al. have investigated the efficacy of omeprazole in the presence of multiple genetic variants in 67 individuals. The rs2860840 variant was associated with decreased concentrations of omeprazole, which might lead to treatment failure; carriers therefore may require higher doses than non-carriers. 40 The rs12248560 variant, also known as CYP2C19 ∗17, has been extensively studied worldwide and confirmed to have ultra-rapid metabolism effects. 41 This variant is significantly associated with in-hospital death among patients with acute coronary syndrome and non-valvular atrial fibrillation who receive clopidogrel with rivaroxaban. Additionally, carriers of the rs12248560 T allele have significantly higher rates of hyper-response to clopidogrel and higher rates of bleeding complications than carriers of the CC genotype. 41 Carriers of the rs12248560 T allele have a significantly higher rate of hyper-response to prasugrel and a higher rate of bleeding complications than carriers of the CC genotype. 42 Furthermore, carriers of the rs12248560 allele might have greater imipramine metabolism than carriers of two normal alleles. Yet, this effect of the rs12248560 allele does not alter the clinical outcomes due to the activity of imipramine's active metabolite, which is desipramine. 43 Nonetheless, carriers of the rs12248560 allele might have greater citalopram metabolism than carriers of two normal alleles. However, conflicting evidence has been reported. 44 Furthermore, carriers of the rs12248560 allele might exhibit greater voriconazole metabolism than carriers of two normal alleles. However, because of uncertainty in the findings, comprehensive local trials are necessary. 45 Several studies have associated the rs11188072 variant with ultra-rapid metabolism of drugs biotransformed by the CYP2C19 enzyme. Carriers of the rs11188072 T allele have an elevated risk of omeprazole treatment failure. 46 Furthermore, carriers of the rs11188072 T allele might have diminished risk of cardiovascular events, because of enhanced clopidogrel response and antiplatelet action. However, a meta-analysis by Bauer et al. has revealed insignificant differences in clopidogrel efficacy among carriers of several rs11188072 genotypes. 47 , 48 In addition, mephenytoin blood concentrations are extremely low among carriers of the rs11188072 TT genotype, in contrast to the wild-type genotype. 46 The rs12769205 variant, also known as CYP2C19 ∗2E or CYP2C19 ∗2J, is significantly associated with diminished metabolism of drugs biotransformed by the CYP2C19 enzyme. Pharmacokinetic studies have indicated that tricyclic antidepressants, including imipramine, clomipramine, trimipramine, amitriptyline, and doxepin, are poorly metabolized, and their blood levels are elevated, among carriers of the rs12769205 mutant allele G (GG or AG). The Clinical Pharmacogenetics Implementation Consortium dosing guideline update recommends considering a 50 % dose reduction in carriers of the rs12769205 mutant allele G. 49 In addition, carriers of the rs12769205 mutant allele G might show a better response to proton pump inhibitors, including lansoprazole, dexlansoprazole, pantoprazole, rabeprazole, esomeprazole, and omeprazole, than carriers of two normal alleles. Hence, a 50 % dose reduction might be considered for carriers of the rs12769205 mutant allele G if proton pump inhibitors are to be used for long periods (>3 months); however, local studies should be conducted to validate this finding. 50 The rs12571421 variant is likely to be associated with poor metabolism of drugs biotransformed by the CYP2C19 enzyme. Carriers of the minor allele of the rs12571421 variant have higher risk of clopidogrel resistance than wild-type allele carriers. 51 In addition, patients taking either citalopram or escitalopram who carry the minor allele of the rs12571421 variant might have better antidepressant efficacy, but a higher risk of adverse events due to poor metabolism, than wild-type allele carriers. 52 Furthermore, HIV-infected patients taking etravirine who carry the minor allele of the rs12571421 variant might have better anti-HIV efficacy, but a higher risk of adverse events due to poor metabolism, than wild-type allele carriers. 53 , 54 Moreover, women who carry the minor allele of the rs12571421 variant might have lower norendoxifen (an aromatase inhibitor and active metabolite of tamoxifen) concentrations, due to poor tamoxifen metabolism, than wild-type allele carriers. 55 No published studies have correlated any disease with the rs12571421 variant. This study has several limitations, notably the omission of the rs4244285 ( CYP2C19 ∗2) variant, whose influence on the efficiency of common medications is well established among carriers. Because the main objective of this study was to determine the frequencies of numerous unmeasured CYP2C19 and CYP2C18 genetic variations in KSA, we focused on examining previously unmeasured or minimally measured variants. Hence, rs4244285 ( CYP2C19 ∗2) was omitted, because its frequency in KSA and globally has been documented. We acknowledge that this omission affected the assessment of complete haplotype analysis and limited the interpretation of associations between the observed CYP2C19 and CYP2C18 variants, particularly those with unconfirmed implications. Nonetheless, the key CYP2C19 ∗2 variants are the two splicing variants rs4244285 and rs12769205, both of which are known to be closely related, to map to the same haplotype, and to be associated with CYP2C19 loss of function. 56 , 57 Hence, the effects of this omission were partially mitigated by the inclusion of rs12769205. Furthermore, several variants were recognized as rare and lacked sufficient statistical power for association analyses, thus restricting interpretability. Additionally, the therapeutic importance of numerous identified polymorphisms affecting the CYP2C19 and CYP2C18 enzymes remains either uncertain or unverified, because in silico functional predictions, including SIFT, CADD, and DANN, have not been experimentally validated. Therefore, to establish an association between a certain genetic variant and particular medications, in vitro validation for high-impact variants should be followed by in vivo studies with a sufficient sample size.

Pharmacogenomics, an important branch of pharmacology, involves the study of genetic variations among diverse populations, including possible effects on drug kinetics and dynamics. Genetic variations among diverse populations are associated with hyper-response, normal-response, hypo-response, non-response, or hypersensitivity to specific drugs. Therefore, pharmacogenomic aspects warrant substantial consideration in medicine. 1 , 2 Numerous mutations in genes that encode metabolic enzymes, such as CYP2C19, influence the pharmacokinetics of multiple medications. CYP2C19, a cytochrome P450 enzyme, oxidizes several widely used medications, such as clopidogrel, tamoxifen, voriconazole, phenytoin, diazepam, escitalopram, progesterone, and omeprazole. 1 , 2 , 3 , 4 , 5 The CYP2C19 enzyme is composed of 490 amino acids and is encoded by the CYP2C19 gene. Mutations in the CYP2C19 gene affect the metabolism of multiple drugs. In addition, because the CYP2C18 and CYP2C19 genes share >80 % sequence identity and are located at the same locus, several genetic mutations of CYP2C1 8 might also influence the metabolism of CYP2C19 enzyme substrates. 6 , 7 , 8 Although multiple drugs are recognized CYP2C19 and CYP2C18 substrates, their degree of metabolism varies depending on the nature of their chemical structures. The CYP2C19 enzyme, and potentially its cognate CYP2C18, are involved in metabolizing several drugs, including clopidogrel, warfarin, lansoprazole, escitalopram, and voriconazole. 7 , 9 The CYP2C19 and CYP2C18 genes have more than 500 known variants; however, many of these variants are rare. 10 The CYP2C19∗1 and CYP2C18∗1 alleles are recognized as wild type and encode normal active forms of CYP2C19 and CYP2C18. Common variants include CYP2C19 ∗2, CYP2C19 ∗3, CYP2C19 ∗17, rs2860840 ( CYP2C18 ), rs2901783 ( CYP2C18) , and rs7917985 ( CYP2C18 ). 11 The CYP2C19∗17 mutant allele encodes a more active version of the CYP2C19 enzyme, whereas the CYP2C19∗2 and CYP2C19∗3 mutant alleles result in the production of an inactive CYP2C19 enzyme. 11 A significantly high prevalence of CYP2C19∗2 and CYP2C19∗3 has been reported in Chinese, Filipino, Thai, Malaysian, Japanese, and Indigenous Australian groups. 12 By contrast, Europeans and Africans have high rates of the CYP2C19∗ 17 allele. 12 Several studies have indicated that Saudi inhabitants of particular regions frequently carry the CYP2C19∗2 , CYP2C19 ∗17, and CYP2C19 ∗1B alleles, whereas very few people carry the CYP2C19∗3 , CYP2C19∗6 , CYP2C19∗8 , CYP2C19∗9 , CYP2C19∗13 , and CYP2C19∗15 alleles. 6 , 11 , 13 In addition, the CYP2C19∗4 , CYP2C19∗7 , CYP2C19∗10 , CYP2C19∗11 , CYP2C19∗12 , and CYP2C19∗14 alleles have not been detected among Saudis. 6 , 13 Previous Saudi investigations were conducted in the cities of Riyadh and Abha, yet limited knowledge is available regarding CYP2C19 variants in other Saudi areas with populations belonging to various Arabic tribes. Additionally, to our knowledge, CYP2C18 genetic variants have not previously been studied in KSA. Although Saudi studies have measured the common CYP2C19 variants, several other variants associated with CYP2C19 and its cognate CYP2C18 gene have not been measured. Hence, this study was intended to fill the data gap regarding other unmeasured CYP2C19 and CYP2C18 genetic variants among inhabitants of KSA, and to investigate the potential clinical and pharmacological implications. Because CYP2C19 and CYP2C18 share a locus and exhibit high sequence homology, they were investigated together. The Alahsa Metropolitan Area was chosen as the study location, because of its diverse population; urban growth due primarily to the presence of lands owned by the Saudi Arabian Oil Company (ARAMCO) and the National Guard; and a substantial population increase from 500,000 inhabitants in 1990 to 1.24 million in 2017. 14

This multicenter cross-sectional study conducted in the Alahsa Metropolitan Area, KSA, from December 2022 to September 2023, involved healthy Saudi adults randomly selected from 22 primary healthcare centers and five governmental hospitals across the four principal cities of the Alahsa region: Al-Hofuf, Al-Mobarraz, Al-Oyoun, and Al-Omran. In Roasoft Inc. software, the minimum sample size was calculated to be 303 participants, on the basis of the estimated adult population in the Alahsa region in 2020 (N = 800,000), a confidence level of 95 %, and an assumed proportion of 0.27 derived from prior studies in KSA suggesting that approximately 27 % of the population has CYP2C19 gene polymorphisms. 10 , 12 Notably, although the calculated sample size was adequate for detecting common variant associations, it was underpowered for detecting rare variant associations (minor allele frequency [MAF] <1 %). Blood samples (4 mL) were collected from each participant and stored at −20 °C before genotype analysis. To extract DNA from the collected blood samples, we used a commercial kit (QIAamp DNA mini kit, Qiagen, USA). After extraction, the DNA was quantified with PicoGreen assays and a Synergy HTX instrument (Promega, cat. E2670). DNA quality and purity were evaluated with a NanoDrop spectrophotometer. Absorbance measurements at 260 nm were used to quantify the DNA. DNA samples with a ratio of absorbance at 260 to 280 nm between 1.7 and 2.0 were considered to be relatively pure and of good quality. These measurements indicated the concentration and purity of the DNA samples. An Infinium global screening array v 3.0 was used for genotyping of markers. According to the manufacturer's instructions, DNA was hybridized to a BeadChip, then scanned with an iScan® system. The bead locations on the BeadChip and quantification of the signal intensity were determined with downloaded BeadChip decode map (DMAP) files. The generated intensity data (∗.idat) files were combined with bead pool manifest (∗.bpm) files containing information on the single-nucleotide polymorphism (SNP)/probe content on the BeadChip and cluster files – also called ensemble graphics toolkit (∗.egt) files – which contain reference information for each interrogated locus. Finally, the these files were combined and converted to genotype call (∗.gtc) files, which were imported into GenomeStudio software for analysis. For data quality control, the GenTrain score, call frequency, number of calls, number of no calls, and ChiTest100 were measured. In addition, GenomeStudio version 2.0 software was used to modify the clusters, normalize data, and call genotypes for all samples. In GenomeStudio, a project was created with ∗.idat files containing the xy location and intensity data. The ∗.bpm file containing the bead ID and sequence information, and the ∗.egt file containing the genotyping cluster positions were used to accelerate the project creation process. These files were imported into GenomeStudio, and the project creation wizard was followed. GenomeStudio was used to evaluate the samples. The main sample metric was the call rate, which is the percentage of all variants and insertion-deletion mutations (indels) successfully genotyped for a given sample. Any sample with a call rate <96 % was excluded. The call rate provided the starting point for quality control of assay or sample performance. The sample metric of logR deviation (logR dev), or the deviation in the normalized intensity value, was used to indicate sample contamination. The main SNP metric was the call frequency; the percentage of samples for which a genotype was called for a specific SNP. Another SNP metric, the GenTrain score, reflected how well GenomeStudio identified groups of alleles in a training data set. An ideal GenTrain score of 1 would have 3 clusters with equal R at 0, 0.5, and 1 norm Theta. Moreover, we used ChiTest100 to calculate the p -value from a chi-squared test of Hardy–Weinberg equilibrium (HWE). The HWE test assesses whether the observed genotype frequencies for a given SNP in the study population are consistent with those expected under HWE. A high p -value indicates a high probability that the observed deviations from expected HWE frequencies are due to random chance, therefore, the data are consistent with the population being in HWE, and the SNP is probably of good quality and accurately genotyped. After samples and SNPs were evaluated, and data quality control was completed, a final report containing the allele calls was exported in variant call format (∗.vcf). For bioinformatics analysis, we used BCFtools in bioinformatics pipelines for various applications, namely, CADD, DANN, regSNP-intron, VEST, and SIFT, to annotate the detected genetic variants. 15 , 16 Furthermore, Haploview version 4.1 was used to analyze the linkage disequilibrium (LD) and HWE between all detected CYP2C19 SNPs among study participants. The constructed haplotypes were visualized through an LD plot generated with Haploview. 17 , 18 To measure LD among the genetic variants, we determined the D′, r 2 , and logarithm of the odds (LOD).

The authors declare no conflicts of interest.

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.