Acute illness encompasses responses to severe infection (sepsis) and injury (trauma), as well as acute pancreatitis and other similar disease states. Acute illness is a complex constellation of responses involving dysregulated inflammatory and immune responses that are ultimately associated with multiple organ dysfunction (1). This highly dynamic process intertwines multiple (and perhaps all) physiologic systems and presents a quintessential complex, dynamic system (2–4). Gene association studies of single-nucleotide polymorphisms (SNPs) have shed extensive insights into and suggested novel biomarkers for both clinical and pharmacological outcomes of various complex diseases (5). However, the application of these genetic analyses has lagged in the context of acute illness, despite fairly extensive studies (6–16). We suggest that one reason these acute illness gene association studies have not yielded more clinically useful insights is due to the lack of standards and clearly defined best practices associated with validation of the purported SNPs.
The importance of validating gene association studies is rooted in clinical application in an era of pharmacogenomics and precision medicine. Disease management was traditionally performed based on identifying the disease and the causative agent and providing therapy. Today, knowledge of biomarkers and gene-based therapies (17), as well as candidate SNP association findings, provide opportunities for personalized disease management. With approximately 4 to 5 million SNPs in the human genome (i.e., gene sequence variations with >1% frequency in the human population (18, 19)), the possibility of spurious SNP-disease associations is a major concern; thus, we suggest that the ultimate clinical utility of assessing particular SNPs will likely depend on validation and quality control (QC) measures. Additionally, the interplay between SNPs and pathology, physiologic responses, and pharmaceuticals all contribute to clinical outcomes.
Illustrating the crucial need for validation in SNP studies, a well-powered replication and validation study of 70 previously published studies of SNPs associated with clinical outcomes after blood or marrow transplantation found only one validated SNP of the 45 SNPs studied. Additionally, these authors found that only 13% of the 45 SNPs were related to gene expression or transcription factor binding. The authors reported the importance of the need for confirming gene association studies by reproducing them in independent samples, with replication of data defined as two studies with similar inclusion criteria and validation of data defined as two studies with different inclusion criteria (20). In addition to replicable and validated data, QC measures for gene association studies are to be specified, such as SNP call rates, discordance between duplicate samples, linkage disequilibrium, control group methods, and/or QC SNPs, further described in Tables 2 and 3(21, 22).
Validation of SNP-disease association studies is therefore important for future application in medicine in general and, as we discuss in this review article, for the field of acute illness in particular. We review clinical and pharmacological gene association studies of SNPs and summarize the SNP validation methods, control groups, and SNP quality measures utilized by these studies as a potential framework for creating a standardized model for gene association SNP validation and quality measurement criteria. Our goal is not to dictate best practices per se, since that is the province of well-established consortia (21, 22). Rather, our goal is to provide examples of the broad array of approaches used to validate conclusions derived from genetic polymorphism studies in other complex diseases to learn lessons applicable to the study of acute illness.
Between December 2018 and January 2019, we searched PubMed for literature from peer-reviewed studies published between 2011 and 2019, reporting data on SNPs and clinical outcomes and SNPs and pharmaceuticals (i.e., pharmacogenomics). This was a focused review of each study's methods section covering SNP quality control measures, control group, and validation (Fig. 1), aimed at providing examples of relevant recent studies, and as such is not intended as an exhaustive review.
We aimed to include studies representing a variety of clinical and pharmacological associations as illustrative examples, and thus this is not a complete list of all gene association studies. Our primary disease process is acute illness for this review. We also cite examples of articles outside the standard acute disease model to provide additional information on how others are approaching the field of SNP association studies. These studies provide unique and interesting examples of SNP research methodology from fields that have carried these types of studies to a greater extent than in the field of acute illness, similar to a recent review on precision medicine in the context of acute illness (4). Our selected studies represent sample publications on SNPs in the context of trauma, sepsis, severe influenza, cancer, diabetes and coronary heart disease, inflammatory bowel disease, age-related macular degeneration, ischemic events after cerebral endovascular therapy, coronary artery bypass surgery. We also included SNP studies from pharmacogenomics in the fields of acute illness, cancer, CABG surgery, and cutaneous reactions to phenytoin.
While we did not conduct a comprehensive list of SNP association studies, we have provided additional references to articles for further reading. This review article is intended to provide information on examples of SNP association studies in acute illness and other diseases, with a focus on the importance of utilizing protocols and SNP validation to readers who are interested in learning more about SNP association studies and or would like to enter this field of research.
When reading the SNP association studies reviewed in this article, we encourage use of Table 2, which outlines an abridged version of The OncoArray guidelines, an abridged protocol with side notes and Table 3 which contains terms from the Nature Protocols study, which are taken directly from their glossary to facilitate optimal understanding (21, 22). We note that the information metric (INFO) score is not listed in the glossary and is a measure of imputation quality, where an INFO score of >0.85 represents optimal variants (23).
VALIDATION IN THE CONTEXT OF CLINICAL GENE ASSOCIATION SNP STUDIES
We first sought to delineate recent progress in the field of acute illness before delving into insights that could be gleaned from assessing the QC measures described in SNP studies in other fields. We note that there is a substantial body of literature associated predominantly with candidate SNP studies in acute illness, reviewed recently (24–27).
Studies in settings of acute illness
Studies over the past 20 years or so have suggested a link between SNPs and outcomes in trauma (8, 28–32) and sepsis (33–38). Below, we focus on recent studies in the arena of critical illness in the context of methodology and validation.
Sepsis in Blunt Trauma: A SNP in the precursor microRNA (miRNA) hsa-mir-608 was found to be associated with sepsis, multiple organ dysfunction, and pro-inflammatory cytokine levels in blunt trauma patients. The screening cohort was n = 666, followed by two independent validation cohorts of n = 286 and n = 316. Methods for identifying SNPs in precursor miRNA and genotyping were discussed. Genotyping was performed in a blinded fashion with approximately 10% of genotyped samples duplicated for QC. SNP QC measures included Hardy-Weinberg equilibrium (HWE) deviation determination and minor allele frequency (MAF) (21, 22). The study's power was discussed, with the screening cohort as 92% and two validation cohorts as 65% and 52% (37).
Immune Response and Sepsis in Trauma: Seaton et al genotyped eight SNPs associated with the melanocortin-1 receptor, an anti-inflammatory mediator possibly involved in the post-traumatic immune response, and found that a specific variant was associated with a lower risk of developing complicated sepsis after injury. SNP genotyping was performed using TaqMan-based real-time polymerase chain reaction (PCR). SNPs with MAF ≤ 1% were excluded. Genome-wide genotyping data on a patient subgroup were available from a previous study and used as “null markers” and then transformed into principal components. QC was performed according to a protocol published in Nature Protocols(22). Samples with gender discordance, excess relatedness, and an autosomal heterozygosity rate of ±3 standard deviations from the mean were excluded. Genetic markers were filtered for call rate < 0.95, MAF < 0.05, and deviations from HWE in the control population (38).
Trauma (Survivors versus Non-Survivors): Seven SNPs were found to be associated with distinguishing trauma survivors from non-survivors, along with a distinct Th17 response between these groups. This retrospective case-control study comprised 13 non-survivors and 384 survivors. Genotyping was performed for all patient samples for 551,839 SNPs. An enrichment strategy was utilized for SNP determination, which consisted of initially comparing SNPs of 13 non-survivors with a control group of 13 matched survivors, producing 126 SNPs. Additional haplotype comparison of the remaining survivors (N = 371) produced seven SNPs. Matching criteria included age, sex ratio, and Injury Severity Score. Serial analysis of inflammatory mediators was performed. Principal component analysis (PCA) and dynamic network analysis was performed on inflammatory mediators (39). In related studies, an SNP in MPPED2 that was one of the seven SNPs was correlated with worse outcomes in severely injured patients compared to propensity-matched, equally severely injured controls (unpublished observations).
Traumatic Brain Injury: A prospective study found SNPs in the sur1 gene (ABCC8) to be associated with increased risk of cerebral edema (CE) in traumatic brain injury in 385 patients. Genotyping methods were discussed, and QC was performed using blind technical duplicates to assess discrepancies; call rate was 95%, MAF >0.2. Research assistants were blinded for genotyping of SNPs and CE outcomes. Linkage disequilibrium was determined. Age, gender, and initial Glasgow Coma Score were included in the analysis to control for confounders (40).
Acute Kidney Injury: Patients with acute kidney injury who had a polymorphism in the catalase gene were found to have increased hospital morbidity and mortality rates. This cross-sectional study included 90 patients and 101 healthy volunteers. Genotyping was performed using PCR-restriction fragment length polymorphism for the manganese superoxide dismutase, catalase, and glutathione peroxidase genes. Genotyping reliability was ensured using two observers; no interobserver variability or discordance was found when 10% of the blinded samples were re-analyzed. HWE testing was included (41).
Sex-Based Differences in Trauma: Sperry et al. performed a prospective, observational cohort study of blunt trauma patients to investigate how an X-chromosome linked SNP of the IL-1 receptor associated kinase (IRAK1) protein may be responsible for sex-based outcome differences following injury. The researchers found that this polymorphism was a strong independent predictor of multiorgan failure and mortality. Genotyping of the SNP was performed using PCR. Allelic discrimination was verified by direct DNA sequencing of a subgroup of patients of each haplotype to assure accuracy of the PCR assay. Multiple confounding variables were controlled for in this study (42).
Aging in Trauma: In (Lamparello et al, Shock, in press) recent unpublished work from our group, a retrospective case-control study was performed in which we identified that an aging-related SNP associated with longevity, rs2075650, may influence clinical outcomes and inflammation biomarker networks in aged patients following blunt trauma. Genomic DNA samples were examined for 551,839 SNPs using a microarray kit. A cohort of aged patients (65–90 years old) with the homozygous major allele genotype of rs2075650 was compared to minor allele carriers. Comparisons were made with a cohort of young patients (18–30 years old) with the same injury severity score as the aged patients to show that the potential impact of the candidate SNP is age-dependent. Additionally, aged patients were stratified according to their genotype of a control SNP, rs5966792, with no differences in clinical outcomes and only one significantly different biomarker between groups (unpublished observations).
We next sought to gain insights into the use of QC measures in SNP studies in fields outside acute illness.
Studies in severe influenza
A SNP associated with disruption of binding of the CTCF transcription factor at the promoter region of IFITM3 (Interferon Induced Transmembrane Protein 3) was found to be linked with the risk of severe influenza. A challenge study of healthy adults (N = 42) was utilized for comparison to two other study groups: a FLU09 naturally acquired influenza cohort (N = 86) and a cohort of critically ill children (N = 265). Mild and severe influenza cases were also assessed. Genotyping was performed using touchdown PCR. Quality control included MAF, concordance, QC samples, linkage disequilibrium determination, as well as Sanger sequencing. Association analysis was found using linear mixed models (43).
Studies in cancer
Gene association studies have indicated a multitude of susceptibility loci with various cancer types and their relative clinical outcomes (44–47). Below, we focus on recent studies in the arena of cancer in the context of methodology and validation.
Breast Cancer: Nine SNPs of matrix metalloproteinases (MMPs), specifically within MMP8 and MMP9, were found to be associated with breast cancer risk in a Chinese Han population of 571 patients and 578 controls. MMP8 and MMP9 SNPs were chosen from previously published literature. Genotyping was discussed and analysis of associations between breast cancer risk and haplotypes was determined. Linkage disequilibrium was performed for MMP8 and MMP9 SNPs (48).
Liver Cancer: Three SNPs were found to be associated with stratifying advanced hepatocellular carcinoma with varying treatment outcomes in a study investigating 116 patients. Genotyping was determined prior to patient chemotherapy treatment. PCR primers for sequencing were provided and SNP determination was found by sequencing in both directions (49).
Ovarian Cancer: Twelve new susceptibility loci were found to be associated with histotypes of epithelial ovarian cancer (EOC) from multiple genome-wide genotyping studies from 25,509 EOC cases and 40,941 controls. Genotyping included OncoArray, the Mayo GWAS (genome-wide association study), the Collaborative Oncological Gene-environment Study (COGS) and others. SNP QC was based on the OncoArray QC guideline. SNPs were excluded with call rates less than 95%. MAF was less than 0.01 and duplicates, close relatives, and concordance were addressed in the study. PCA was utilized for the OncoArray data (50).
Laryngeal Cancer: An O6-methylguanine DNA methyltransferase (MGMT) SNP was found to be associated with the incidence of laryngeal squamous cell carcinoma in 96 patients and 102 control participants. Microsequencing was utilized to analyze p53 and MGMT polymorphism, and four polymorphic sites were chosen from The Single Nucleotide Polymorphism Database and Haplotypemap dataset. SNP selection was confirmed with prior tumor association studies. Genotyping methods were discussed and linkage disequilibrium and HWE deviation were determined (51).
Leukemia, Breast, and Lung Cancers: We assessed the first known study to associate the cytochrome P450 CYP3A7*1c allele with cancer outcomes in patients. Utilizing the effects of CYP3A on urinary estrone glucuronide levels, the study investigated the CYP3A genotype on cancer outcomes in chronic lymphocytic leukemia, breast, and lung cancers. Genotyping for breast cancer patients in a previous study was performed with customized bead arrays and competitive allele-specific PCR. Estrone glucuronide level associations with the CYP3A7*1 allele were performed using CEU 1000 genome pilot data. The study designed Sequenom plexes and had seven SNPs for QC. Additional QC measures were call rates listing a mean of 99.4%, utilization of HWE deviation determination, 100% duplicate sample concordance, and INFO score. For a SNP to be found significant, confirmation of imputed genotypes and association with outcome was performed with listed call rate and concordance between duplicate samples determined. Sanger sequencing was utilized to confirm SNP proxy for the CYP3A7*1 allele (52).
Lung Cancer (Adenocarcinoma): SNPs in the phosphatase and tensin homolog (PTEN) gene were found to be associated with advanced lung adenocarcinoma in 618 patients receiving platinum-cased chemotherapy. A SNP database was used to find the gene region of interest for PTEN with four SNPs identified using tagger algorithm and genotyping performed using genotyping assays. Approximately 15% of samples were randomly chosen for repeat genotyping with concordant results. QC excluded SNPs with call rates less than 95%, MAF <0.05, and HWE deviation <0.05 with linkage disequilibrium determined (53).
Colorectal Cancer: Polymorphisms in cytochrome P450, specifically the CYP3A5 gene, were found to be associated with progress-free survival in patients with metastatic colorectal cancer who received irinotecan, 5-fluorouracil, and leucovorin (FOLFIRI) combination chemotherapy. This prospective study included 82 patients with 79 SNPs selected from the HapMap Project database from a prior study. SNPs were genotyped with matric-assisted laser desorption/ionization time-of-flight mass spectrometry or with PCR. Quality measures included MAF and HWE deviation determination (54).
Lung Cancer (NSCLC): Investigators genotyped 240 miRNA-related SNPs in 535 patients with stage I and II non-small cell lung cancer (NSCLC) and found associations between miRNA-related polymorphisms and clinical outcomes, specifically mortality, disease recurrence, and survival. Genotyping was performed using a custom genotyping platform. SNP call rates >95% were included as well as duplicates for 2% of samples. Concordance was >99%, MAF was >0.01, and linkage disequilibrium was determined. Control cells were utilized in luciferase reporter assay. Internal validation of results was performed using bootstrap resampling analysis (55).
Leukemia: SNPs associated with the risk of developing chronic lymphocytic leukemia (CLL) are lower in African Americans compared to Caucasians. This study had a discovery cohort of 42 African American patients with CLL, which was confirmed by a second cohort of 68 African American patients with CLL from the CLL Research Consortium. Eight ancestry-informative SNPs and 15 CLL risk SNPs were genotyped with SNP genotyping assays. SNP call rates >95% were included, a minimum MAF >0.60 was utilized, and concordance was determined between two CLL groups. Ancestry SNPs were selected by HapMap allele frequencies. Linkage disequilibrium was determined. CLL risk alleles were found from previously published data on Caucasian patients. Control African American allele frequencies (N = 530) were found from genome-wide association data and the HapMap database (56).
Studies in diabetes and coronary heart disease
Studies in vasculature, insulin resistance, leukocyte migration, inflammation, adiposity, and others have been linked with susceptibility loci and clinical outcomes in heart disease (57–62). Below, we focus on a recent study in the arena of coronary heart disease and diabetes associated with SNPs in the context of methodology and validation.
The authors who obtained data from 265,678 participants found 17 loci associated with type 2 diabetes (T2D), a locus associated with coronary heart disease (CHD) and identified shared loci between T2D and CHD. Additionally, common pathways found in the study suggest new potential therapeutic targets. The controls for the study were patient-based, with 192,341 control participants for the T2D loci discovery and 169,534 control participants for outcomes connected with both T2D and CHD. Genotyping was performed using a high-density genotyping array and genotyping QC was provided in supplemental tables. Call rates, variant rates, and gender matching were determined, and exclusion of variants was based on MAF, INFO score, and maximum posterior call. Population stratification was assessed using PCA (63).
Studies in inflammatory bowel disease
In 2017, 215 risk loci were associated with inflammatory bowel disease, with ongoing new loci being discovered as well as risk score identification, inherited determinants, race-specific associations, and prognosis (64–68). Below, we discuss a recent study focusing on risk loci in the arena of inflammatory bowel disease associated with SNPs in the context of methodology and validation.
A GWAS found 25 risk loci for inflammatory bowel disease in a sample of 59,957 subjects with 13,145 population controls. Genotyping was performed and GWAS QC excluded variants meeting the following criteria: no overlap between both versions of the chip, missingness >5%, call rate difference between case and control 1P<1 × 10−5, deviation from HWE, and genotyping batch effect in outliers using PCA. Association testing was performed with an additive frequentist association test. Variants were removed with MAF <0.1%, INFO <0.4. Linkage disequilibrium was calculated and European linkage disequilibrium scores were utilized. Fine mapping analysis was performed to identify causality between loci and outcome (64).
Studies in age-related macular degeneration
Several studies of susceptibility loci associated with regulation of compliment factors, mononuclear phagocyte activity, and vasculopathy have been linked with age-related macular degeneration (AMD) (69–73). Below, we focus on a recent study in the arena of AMD associated with SNPs and neovascularization in the context of methodology and validation.
SNPs associated with the CFH, ARMS2, and C3 genes are associated with AMD and features of neovascularization. Patients for this study were recruited from the Comparison of Age-Related Macular Degeneration Treatments Trials (CATT), a multicenter, single-blind, noninferiority randomized trial of patients receiving injections of ranibizumab or bevacizumab. From the CATT study, 835 patients were genotyped for SNPs associated with AMD using custom-made genotyping assays. Linkage disequilibrium was determined. Age, gender, and smoking status were adjusted in patient analysis (74).
Studies in the setting of ischemic events after cerebral endovascular therapy
Susceptibility loci have been implicated with the risk of stroke and clopidogrel efficacy, central adiposity, and folic acid intervention (34, 57, 75, 76). Below, we focus on a recent study in the arena of ischemic events in endovascular therapy and CYP2C19*17 polymorphism in the context of methodology and validation.
A prospective cohort study of 108 patients who had endovascular therapy and had the CYP2C19*17 polymorphism was found to be associated with increased risk of ischemic events, which were not dependent on clopidogrel response. Genotyping was performed using allele-specific PCR analysis and TaqMan-based real-time PCR from previous studies. Four exclusive genotype groups were created and analyzed to determine the impact of CYP2C19 polymorphisms. Group 1 comprised wild-type carriers, who also were the control group N = 44 (77).
Studies in coronary artery bypass surgery
Susceptibility loci have been implicated with all-cause mortality associated with chromosome 9p21, interleukin-6 promoter gene variant with inflammation and atrial fibrillation, and nitric acid synthase gene variant with vascular responsiveness to phenylephrine (78–80). Below, we focus on a recent study in the arena of coronary artery bypass surgery and susceptibility loci in the thrombomodulin gene in the context of methodology and validation.
Allele variants in the thrombomodulin gene were found to be associated with increased long-term mortality after coronary artery bypass surgery (CABG). The study included two independent CABG cohort groups that consisted of a discovery cohort (N=1018) and a validation cohort (N=930). All-cause mortality between 30 days and 5 years was measured. Genotyping in the discovery cohort was performed using matric-assisted laser desorption/ionization time-of-flight mass spectrometry on a Sequenom system. Genotyping in the validation cohort was performed using a bead chip. For the discovery group, analysis included Sequenom data for intensity plots and genotype calls. For the validation group, separate software was utilized for raw data. Genotype QC criteria included genotype calls and MAF >5%. An ABI 3700 capillary sequencer was used for genotyping accuracy, in which six SNPs were scored out of 100 random patients. Linkage disequilibrium calculation was performed (81).
VALIDATION IN THE CONTEXT OF PHARMACOGENOMICS
Studies in settings of acute illness
Compared to relatively simpler studies of gene polymorphisms in acute illness outcomes, the interaction between specific SNPs and drugs (pharmacogenomics) in acute illness has been studied to a much lesser degree. Below, we summarize recent studies in this field.
Sepsis and Linezolid: SNPs associated with linezolid elimination were associated with sepsis outcomes based on a score in 14 ICU-admitted patients receiving intravenous linezolid (600 milligrams every 12 h). The sepsis scores generated by the authors included group 0 for sepsis, group 1 for severe sepsis, and group 2 for septic shock. Genotyping was performed using a real-time PCR allelic discrimination assay. SNP QC measures included HWE deviation determination and SNP exclusion based on MAF. Duplicate analysis was performed with no discrepancy in data results, and linkage disequilibrium was calculated (82).
Trauma, Sevoflurane, and Propofol: In recent unpublished work, we found SNPs previously associated with sevoflurane to be linked with worse clinical outcomes in blunt trauma patients. Additionally, the sevoflurane-related SNPs and propofol-related SNPs were linked with altered inflammatory markers. Genomic DNA samples were examined for 551,839 SNPs, and 31 inflammatory biomarkers were analyzed. Confounders were controlled for based on seven criteria. Genotyping methods were discussed. SNP QC measures included call rates <99%; duplicates were removed, HWE was demonstrated, and linkage disequilibrium was determined (unpublished observations). Limitations to SNP analysis in our study were finding a different neighboring SNP one base-pair away from one of our candidate SNPs, making sequence confirmation challenging.
Studies in cancer
Colorectal Cancer and NSAIDs: The regular use of aspirin or non-steroidal anti-inflammatory drugs (NSAIDs) was found to be associated with a lower colorectal cancer risk based on genomic differences at two SNPs on chromosomes 12 and 15. This was a case-control study that enrolled colorectal cancer patients and matched controls between 1976 and 2003. There were 8,634 participants and 8,553 controls matched 1:1 in a time-forward manner. Age, gender, medical facility, and race were adjusted for in analysis. Genotyping was performed with genome-wide association scans. Genotyping samples were excluded based on call rates <98%, heterozygosity, unexpected duplicates, gender, unexpected high identity-by-descent, or unexpected genotype concordance (>65%) with another individual. Linkage disequilibrium was determined using CEU population data. Additionally, SNPs were excluded if <5% MAF, deviation from HWE in control samples, triallelic, no assigned rs number, or inconsistent performance across platforms (83).
Pediatric Leukemia and Vincristine: A study of children with acute lymphoblastic leukemia (ALL) who had a variant of centrosomal protein 72 (CEP72) and received vincristine treatment had an increased risk and severity of vincristine-related peripheral neuropathy. Participants from two prospective ALL clinical trials were enrolled, and 321 patients had DNA available for genome-wide SNP analysis. Human leukemia cells and induced pluripotent stem cell neurons were used. Genotyping methods were discussed. SNPs were excluded based on call rates <95% and MAF <1%. Genetic ancestry was analyzed, and control RNA was mentioned in the supplemental methods section (84).
Studies in the setting of CABG and β-adrenoceptor blockers (βAR blockers)
GNAS (guanine nucleotide binding protein, alpha stimulating) gene variants were found to be associated with an increased risk of mortality in patients of European ancestry who received βAR blockers and had primary CABG surgery. This prospective study analyzed 1,627 patients who were genotyped for genetic variants of GNAS and all-cause mortality for up to 5 years was included. Genotyping was performed using “slow down PCR” and SNP QC measures included HWE (85).
Studies in the setting of cutaneous reactions and phenytoin
A genetic variant of cytochrome P450, specifically CYP2C9*3, was found to be associated with severe cutaneous reactions related to phenytoin use. In this case-control study of 105 patients with phenytoin-related cutaneous adverse reactions and 3,655 patient controls, a GWAS was performed and results in three separate groups were validated. Genotyping methods were discussed, and genotype calls were calculated with a mean call rate of 98.7% and exclusion of SNPs with call rates less than 0.90. MAF was utilized in this study. PCA and HWE deviation determination was utilized. Functional SNPs were determined using PCR and further examination was performed. Linkage disequilibrium was determined (86).
In this review article, we have found three key points that strengthen the potential accuracy of SNP association studies in acute illness and other diseases: providing evidence of following a protocol quality control method such as the one in Nature Protocols(22) or the OncoArray QC Guidelines (21); enrolling sufficient patients to have large cohort groups; and validating the SNPs using an independent technique such as a second study using the same SNPs with new patient cohorts. We chose to focus on gene-association studies of SNPs based on the model that injury severity, elicited inflammatory response, and inherent genomics are variables predictive of clinical outcomes (Fig. 1). Given this finding of inherent genomics as a variable predictive of clinical outcomes, we focused on inherent genomics by reviewing selected examples of clinical and pharmacological gene-association studies of SNPs. Below is a summary of findings from the reviewed studies (see Table 1 for full details).
Control Group and Genotyping: When reviewing presence of a control group, studies were found to have patient control groups with QC SNPs, enrichment strategy, wild-type carriers, control RNA, and control cells; 10 studies did not list a control group. Control group participants ranged from 13 to 192,341 patients. Genotyping methods and QC measures including ancestry determination and matching were items shared in the methods sections for control group data. All studies listed their genotyping methods, and some studies also listed PCR primer sequences. Methods or supplemental data sections provided types of genotyping and SNP QC measures.
SNP QC Measures: Common QC measures found in reviewed studies were call rates, MAF, and HWE deviation determination. Less common QC measures found in studies were exclusion of duplicates and concordance determination, INFO score, ancestry, gender, heterozygosity, variants not associated with the 1000 genome project, and PCA. Given a trend in utilizing QC measures, a standardized protocol for QC measures may be the next step in optimal processing of data in gene association studies. Using known guidelines such as The OncoArray QC guidelines and the genetic case-control association study protocols published in Nature Protocols(21, 22) (Tables 2 and 3) may provide a platform on which to facilitate a global discussion or create unified QC measures for gene association studies of SNPs (Table 1).
Validation: Validation methods varied in the studies we reviewed, and most studies suggest that the next step is validation in a larger cohort with sufficient discriminative power. One trauma study looked at survivors versus non-survivors initiated with a small control group of 13 followed by an enrichment strategy that yielded significant SNP associations with trauma (39). Another study of phenytoin-related cutaneous reactions had the initial group, which found significant SNPs, and followed with three validation groups (86). A third study investigating clinical outcomes in early-stage NSCLC patients used internal validation with bootstrap re-sampling analysis for validation (55). A study looking at the associations between SNPs and CLL in African American patients utilized a discovery cohort (N = 530) and a smaller validation cohort (N = 68) (56). Similarly, a study of thrombomodulin gene variants associated with mortality in CABG surgery utilized a discovery cohort (N = 1018) and a validation cohort (N = 930) (81). A trauma study in sepsis had a screening cohort (N = 666) and two independent validation groups (N = 286 and N = 316) (37) (Table 1).
Three key points found in our review of SNP association studies in acute illness and other diseases are discussed below.
SNP association studies should provide evidence of following a protocol quality control method such as the one in Nature Protocols(22) or the OncoArray QC Guidelines (21), (Tables 2 and 3). SNP association articles aim and are encouraged to disclose items from their protocol (s) such as SNP call rates, genotyping methods, elimination of bias DNA samples, linkage equilibrium, and other items listed in the protocol(s).
SNP association studies should enroll a sufficient number of patients to have large cohort groups: Large patient cohort sizes increase the sensitivity of study and the study's power, reducing the chance for a Type II error. When planning the study, enlisting large control groups further enhances the quality of the study and enables us to also plan for validation as described below.
Large patient cohort sizes can also help overcome bias based on ethnicity, sex/gender, and other factors. SNP association studies ideally should validate the SNPs using an independent technique: Ideally, using cohorts with new patients to validate the same SNP(s) with the disease of interest will facilitate the accuracy of the association when combined with following SNP protocol guidelines as listed above.
The examples of SNP association studies provided in this review article and references to sample protocols with term definitions and other materials, we hope to provide readers a guide on better understanding the methodology entailed in SNP association studies by showcasing examples of SNP association studies and providing references to sample protocols with term definitions and other materials (23, 87, 88).
The studies presented herein hopefully demonstrate the value of quality control methods in SNP association studies. We illustrate this point with examples from both acute illness and other disease contexts, with the goal of providing a quick reference framework for investigators in the trauma and sepsis fields. It is our hope that this brief review will serve the shock research community both as a condensed reference and as a compendium of recent references from diverse fields.
1. Efron PA, Mohr MA, Bihorac A, Horiguchi H, Hollen MK, Segal MS, Baker HV, Leeuwenburgh C, Moldawer LL, Moore FA, et al. Persistent inflammation, immunosuppression, and catabolism and the development of chronic critical illness after surgery. Surgery
2. Day JD, Cockrell C, Namas R, Zamora R, An G, Vodovotz Y. Inflammation and disease: modelling and modulation of the inflammatory response to alleviate critical illness. Curr Opin Syst Biol
3. Seymour CW, Gomez H, Chang CH, Clermont G, Kellum JA, Kennedy J, Yende S, Angus DC. Precision medicine for all
? Challenges and opportunities for a precision medicine approach to critical illness. Crit Care
21 (1):257, 2017.
4. Buchman TG, Billiar TR, Elster E, Kirk AD, Rimawi RH, Vodovotz Y, Zehnbauer BA. Precision medicine for critical illness and injury. Crit Care Med
44 (9):1635–1638, 2016.
5. MacKenzie M, Hall R. Pharmacogenomics and pharmacogenetics for the intensive care unit: a narrative review. Can J Anaesth
64 (1):45–64, 2017.
6. Chung TP, Laramie JM, Province M, Cobb JP. Functional genomics of critical illness and injury. Crit Care Med
30: (1 suppl): S51–S57, 2002.
7. Lin MT, Albertson TE. Genomic polymorphisms in sepsis. Crit Care Med
32 (2):569–579, 2004.
8. Cobb JP, O’Keefe GE. Injury research in the genomic era. Lancet
363 (9426):2076–2083, 2004.
9. Stuber F, Klaschik S, Lehmann LE, Schewe JC, Weber S, Book M. Cytokine promoter polymorphisms in severe sepsis. Clin Infect Dis
41: (suppl 7): S416–S420, 2005.
10. Dahmer MK, Randolph A, Vitali S, Quasney MW. Genetic polymorphisms in sepsis. Pediatr Crit Care Med
6: (3 suppl): S61–73, 2005.
11. Papathanassoglou ED, Giannakopoulou MD, Bozas E. Genomic variations and susceptibility to sepsis. AACN Adv Crit Care
17 (4):394–422, 2006.
12. Winkelman C. Inflammation and genomics in the critical care unit. Crit Care Nurs Clin North Am
20 (2):213–221, 2008.
13. Wheeler DS, Zingarelli B, Wheeler WJ, Wong HR. Novel pharmacologic approaches to the management of sepsis: targeting the host inflammatory response. Recent Pat Inflamm Allergy Drug Discov
3 (2):96–112, 2009.
14. Reddy AJ, Kleeberger SR. Genetic polymorphisms associated with acute lung injury. Pharmacogenomics
10 (9):1527–1539, 2009.
15. Wong HR. Genetics and genomics in pediatric septic shock. Crit Care Med
40 (5):1618–1626, 2012.
16. Abraham E. It's all
in the genes: moving toward precision medicine in critical illness. Crit Care Med
41 (5):1363–1364, 2013.
17. Jameson JL, Longo DL. Precision medicine: personalized, problematic, and promising. N Engl J Med
372 (23):2229–2234, 2015.
18. Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA, Abecasis GR. A global reference for human genetic variation. Nature
526 (7571):68–74, 2015.
19. Help Me Understand Genetics page: National Library of Medicine (US). Genetics Home Reference [Internet]. Bethesda (MD): The Library; 2019 June. What are single nucleotide polymorphisms (SNPs)?; [cite 2018 Nov 1]; [about 2 screens]. Available at: https://ghr.nlm.nih.gov/primer/genomicresearch/snp
. Accessed June 5, 2019.
20. Karaesmen E, Rizvi AA, Preus LM, McCarthy PL, Pasquini MC, Onel K, Zhu X, Spellman S, Haiman CA, Stram DO, et al. Replication and validation
of genetic polymorphisms associated with survival after allogeneic blood or marrow transplant. Blood
130 (13):1585–1596, 2017.
21. Amos CI, Dennis J, Wang Z, Byun J, Schumacher FR, Gayther SA, Casey G, Hunter DJ, Sellers TA, Gruber SB, et al. The OncoArray Consortium: a network for understanding the genetic architecture of common cancers. Cancer Epidemiol Biomarkers Prev
26 (1):126–135, 2017.
22. Anderson CA, Pettersson FH, Clarke GM, Cardon LR, Morris AP, Zondervan KT. Data quality control
in genetic case-control association studies. Nat Protoc
5 (9):1564–1573, 2010.
23. Coleman JR, Euesden J, Patel H, Folarin AA, Newhouse S, Breen G. Quality control
, imputation and analysis of genome-wide genotyping data from the Illumina HumanCoreExome microarray. Brief Funct Genomics
15 (4):298–304, 2016.
24. Zhang AQ, Yue CL, Gu W, Du J, Wang HY, Jiang J. Association between CD14 promoter -159C/T polymorphism
and the risk of sepsis and mortality: a systematic review and meta-analysis. PLoS One
8 (8):e71237, 2013.
25. Liu R, Mo YY, Wang HL, Tan Y, Wen XJ, Deng MJ, Yan H, Li L. The relationship between toll like receptor 4 gene rs4986790 and rs4986791 polymorphisms and sepsis susceptibility: A meta-analysis. Sci Rep
26. Gao JW, Zhang AQ, Pan W, Yue CL, Zeng L, Gu W, Jiang J. Association between IL-6-174G/C polymorphism
and the risk of sepsis and mortality: a systematic review and meta-analysis. PLoS One
10 (3):e0118843, 2015.
27. Namath A, Patterson AJ. Genetic polymorphisms in sepsis. Crit Care Clin
25 (4):835–856, 2009.
28. De Maio A, Torres MB, Reeves RH. Genetic determinants influencing the response to injury, inflammation, and sepsis. Shock
23 (1):11–17, 2005.
29. Hildebrand F, Pape HC, van Griensven M, Meier S, Hasenkamp S, Krettek C, Stuhrmann M. Genetic predisposition for a compromised immune system after multiple trauma
24 (6):518–522, 2005.
30. Giannoudis PV, van GM, Tsiridis E, Pape HC. The genetic predisposition to adverse outcome after trauma
. J Bone Joint Surg Br
89 (10):1273–1279, 2007.
31. Hildebrand F, Mommsen P, Frink M, van Griensven M, Krettek C. Genetic predisposition for development of complications in multiple trauma
35 (5):440–448, 2011.
32. Bronkhorst MW, Patka P, Van Lieshout EM. Effects of sequence variations in innate immune response genes on infectious outcome in trauma
patients: a comprehensive review. Shock
44 (5):390–396, 2015.
33. Cader MZ, Boroviak K, Zhang Q, Assadi G, Kempster SL, Sewell GW, Saveljeva S, Ashcroft JW, Clare S, Mukhopadhyay S, et al. C13orf31 (FAMIN) is a central regulator of immunometabolic function. Nat Immunol
17 (9):1046–1056, 2016.
34. Chen BB, Coon TA, Glasser JR, McVerry BJ, Zhao J, Zhao Y, Zou C, Ellis B, Sciurba FC, Zhang Y, et al. A combinatorial F box protein directed pathway controls TRAF adaptor stability to regulate inflammation. Nat Immunol
14 (5):470–479, 2013.
35. Cook DN, Pisetsky DS, Schwartz DA. Toll-like receptors in the pathogenesis of human disease. Nat Immunol
5 (10):975–979, 2004.
36. Savva A, Brouwer MC, Roger T, Valls Seron M, Le Roy D, Ferwerda B, van der Ende A, Bochud PY, van de Beek D, Calandra T. Functional polymorphisms of macrophage migration inhibitory factor as predictors of morbidity and mortality of pneumococcal meningitis. Proc Natl Acad Sci U S A
113 (13):3597–3602, 2016.
37. Zhang AQ, Gu W, Zeng L, Zhang LY, Du DY, Zhang M, Hao J, Yue CL, Jiang J. Genetic variants of microRNA sequences and susceptibility to sepsis in patients with major blunt trauma
. Ann Surg
261 (1):189–196, 2015.
38. Seaton ME, Parent BA, Sood RF, Wurfel MM, Muffley LA, O’Keefe GE, Gibran NS. Melanocortin-1 receptor polymorphisms and the risk of complicated sepsis after trauma
: a candidate gene association study. Shock
47 (1):79–85, 2017.
39. Schimunek L, Namas RA, Yin J, Liu D, Barclay D, El-Dehaibi F, Abboud A, Lindberg H, Zamora R, Billiar TR, et al. An enrichment strategy yields seven novel single nucleotide polymorphisms associated with mortality and altered Th17
responses following blunt trauma
49 (3):259–268, 2018.
40. Jha RM, Puccio AM, Okonkwo DO, Zusman BE, Park SY, Wallisch J, Empey PE, Shutter LA, Clark RS, Kochanek PM, et al. ABCC8
single nucleotide polymorphisms are associated with cerebral edema
in severe TBI. Neurocrit Care
26 (2):213–224, 2017.
41. Kidir V, Uz E, Yigit A, Altuntas A, Yigit B, Inal S, Uz E, Sezer MT, Yilmaz HR. Manganese superoxide dismutase, glutathione peroxidase and catalase gene polymorphisms and clinical outcomes in acute kidney injury. Ren Fail
38 (3):372–377, 2016.
42. Sperry JL, Zolin S, Zuckerbraun BS, Vodovotz Y, Namas R, Neal MD, Ferrell RE, Rosengart ME, Peitzman AB, Billiar TR. X chromosome-linked IRAK-1 polymorphism
is a strong predictor of multiple organ failure and mortality postinjury. Ann Surg
260 (4):698–703, 2014.
43. Allen EK, Randolph AG, Bhangale T, Dogra P, Ohlson M, Oshansky CM, Zamora AE, Shannon JP, Finkelstein D, Dressen A, et al. SNP
-mediated disruption of CTCF
binding at the IFITM3
promoter is associated with risk of severe influenza in humans. Nat Med
23 (8):975–983, 2017.
44. Boutros PC, Fraser M, Harding NJ, de Borja R, Trudel D, Lalonde E, Meng A, Hennings-Yeomans PH, McPherson A, Sabelnykova VY, et al. Spatial genomic heterogeneity within localized, multifocal prostate cancer. Nat Genet
47 (7):736–745, 2015.
45. Martincorena I, Raine KM, Gerstung M, Dawson KJ, Haase K, Van Loo P, Davies H, Stratton MR, Campbell PJ. Universal patterns of selection in cancer and somatic tissues. Cell
171 (5):1029, 2017.
46. McKay JD, Hung RJ, Han Y, Zong X, Carreras-Torres R, Christiani DC, Caporaso NE, Johansson M, Xiao X, Li Y, et al. Large-scale association analysis identifies new lung cancer susceptibility loci and heterogeneity in genetic susceptibility across histological subtypes. Nat Genet
49 (7):1126–1132, 2017.
47. Stracquadanio G, Wang X, Wallace MD, Grawenda AM, Zhang P, Hewitt J, Zeron-Medina J, Castro-Giner F, Tomlinson IP, Goding CR, et al. The importance of p53 pathway genetics in inherited and somatic cancer genomes. Nat Rev Cancer
16 (4):251–265, 2016.
48. Wang K, Zhou Y, Li G, Wen X, Kou Y, Yu J, He H, Zhao Q, Xue F, Wang J, et al. MMP8 and MMP9 gene polymorphisms were associated with breast cancer risk in a Chinese Han population. Sci Rep
8 (1):13422, 2018.
49. Lin WR, Hsu CW, Yeh CS, Chen YC, Chang ML, Liang KH, Lin CC, Chu YD, Yeh CT. Combinations of single nucleotide polymorphisms WWOX-rs13338697, GALNT14-rs9679162 and rs6025211 effectively stratify outcomes of chemotherapy in advanced hepatocellular carcinoma. Asia Pac J Clin Oncol
14 (2):e54–e63, 2018.
50. Phelan CM, Kuchenbaecker KB, Tyrer JP, Kar SP, Lawrenson K, Winham SJ, Dennis J, Pirie A, Riggan MJ, Chornokur G, et al. Identification of 12 new susceptibility loci for different histotypes of epithelial ovarian cancer
. Nat Genet
49 (5):680–691, 2017.
51. Lv Y, Jia C, Jiang A, Zhang H, Wang Y, Liu F, Yang L, Sun Y, Lv R, Song X. Analysis of association between MGMT
and p53 gene single nucleotide polymorphisms and laryngeal cancer. Anticancer Res
37 (8):4399–4403, 2017.
52. Johnson N, De Ieso P, Migliorini G, Orr N, Broderick P, Catovsky D, Matakidou A, Eisen T, Goldsmith C, Dudbridge F, et al. Cytochrome P450 Allele CYP3A7*1C associates with adverse outcomes in chronic lymphocytic leukemia
, breast, and lung cancer. Cancer Res
76 (6):1485–1493, 2016.
53. Yang Y, Xu W, Liu D, Ding X, Su B, Sun Y, Gao W. PTEN
polymorphisms contribute to clinical outcomes of advanced lung adenocarcinoma patients treated with platinum-based chemotherapy. Tumour Biol
37 (6):7785–7796, 2016.
54. Dong N, Meng F, Wu Y, Wang M, Cui Y, Zhang S. Genetic polymorphisms in cytochrome P450 and clinical outcomes of FOLFIRI
chemotherapy in patients with metastatic colorectal cancer. Tumour Biol
36 (10):7691–7698, 2015.
55. Pu X, Roth JA, Hildebrandt MA, Ye Y, Wei H, Minna JD, Lippman SM, Wu X. MicroRNA-related genetic variants associated with clinical outcomes in early-stage non-small cell lung cancer
patients. Cancer Res
73 (6):1867–1875, 2013.
56. Coombs CC, Rassenti LZ, Falchi L, Slager SL, Strom SS, Ferrajoli A, Weinberg JB, Kipps TJ, Lanasa MC. Single nucleotide polymorphisms and inherited risk of chronic lymphocytic leukemia
among African Americans. Blood
120 (8):1687–1690, 2012.
57. Dale CE
, Fatemifar G, Palmer TM, White J, Prieto-Merino D, Zabaneh D, Engmann JEL, Shah T, Wong A, Warren HR, et al. Causal associations of adiposity and body fat distribution with coronary heart disease
, stroke subtypes, and Type 2 diabetes
mellitus: a mendelian randomization analysis. Circulation
135 (24):2373–2388, 2017.
58. Howson JMM, Zhao W, Barnes DR, Ho WK, Young R, Paul DS, Waite LL, Freitag DF, Fauman EB, Salfati EL, et al. Fifteen new risk loci for coronary artery disease highlight arterial-wall-specific mechanisms. Nat Genet
49 (7):1113–1119, 2017.
59. Klarin D, Zhu QM, Emdin CA, Chaffin M, Horner S, McMillan BJ, Leed A, Weale ME, Spencer CCA, Aguet F, et al. Genetic analysis in UK Biobank links insulin resistance and transendothelial migration pathways to coronary artery disease. Nat Genet
49 (9):1392–1397, 2017.
60. Swerdlow DI, Holmes MV, Kuchenbaecker KB, Engmann JE, Shah T, Sofat R, Guo Y, Chung C, Peasey A, Pfister R, et al. The interleukin-6 receptor as a target for prevention of coronary heart disease
: a mendelian randomisation analysis. Lancet
379 (9822):1214–1224, 2012.
61. Larsson SC, Burgess S, Michaelsson K. Association of genetic variants related to serum calcium levels with coronary artery disease and myocardial infarction. JAMA
318 (4):371–380, 2017.
62. Gupta RM, Hadaya J, Trehan A, Zekavat SM, Roselli C, Klarin D, Emdin CA, Hilvering CRE, Bianchi V, Mueller C, et al. A genetic variant associated with five vascular diseases is a distal regulator of endothelin-1 Gene Expression. Cell
170 (3):522–533e15, 2017.
63. Zhao W, Rasheed A, Tikkanen E, Lee JJ, Butterworth AS, Howson JMM, Assimes TL, Chowdhury R, et al. Identification of new susceptibility loci for type 2 diabetes
and shared etiological pathways with coronary heart disease
. Nat Genet
49 (10):1450–1457, 2017.
64. de Lange KM, Moutsianas L, Lee JC, Lamb CA, Luo Y, Kennedy NA, Jostins L, Rice DL, Gutierrez-Achury J, Ji SG, et al. Genome-wide association study
implicates immune activation of multiple integrin genes in inflammatory bowel disease. Nat Genet
49 (2):256–261, 2017.
65. Marigorta UM, Denson LA, Hyams JS, Mondal K, Prince J, Walters TD, Griffiths A, Noe JD, Crandall WV, Rosh JR, et al. Transcriptional risk scores link GWAS
to eQTLs and predict complications in Crohn's disease. Nat Genet
49 (10):1517–1521, 2017.
66. Cleynen I, Boucher G, Jostins L, Schumm LP, Zeissig S, Ahmad T, Andersen V, Andrews JM, Annese V, Brand S, et al. Inherited determinants of Crohn's disease and ulcerative colitis phenotypes: a genetic association study. Lancet
387 (10014):156–167, 2016.
67. Brant SR, Okou DT, Simpson CL, Cutler DJ, Haritunians T, Bradfield JP, Chopra P, Prince J, Begum F, Kumar A, et al. Genome-wide association study
identifies African-Specific susceptibility loci in African Americans with inflammatory bowel disease. Gastroenterology
152 (1):206–217e2, 2017.
68. Lee JC, Biasci D, Roberts R, Gearry RB, Mansfield JC, Ahmad T, Prescott NJ, Satsangi J, Wilson DC, Jostins L, et al. Genome-wide association study
identifies distinct genetic contributions to prognosis and susceptibility in Crohn's disease. Nat Genet
49 (2):262–268, 2017.
69. Calippe B, Augustin S, Beguier F, Charles-Messance H, Poupel L, Conart JB, Hu SJ, Lavalette S, Fauvet A, Rayes J, et al. Complement factor H
inhibits CD47-mediated resolution of inflammation. Immunity
46 (2):261–272, 2017.
70. Yates JR, Sepp T, Matharu BK, Khan JC, Thurlby DA, Shahid H, Clayton DG, Hayward C, Morgan J, Wright AF, et al. Complement C3
variant and the risk of age-related macular degeneration
. N Engl J Med
357 (6):553–561, 2007.
71. Mattapallil MJ, Caspi RR. Compliments of factor H: what's in it for AMD
46 (2):167–169, 2017.
72. Fritsche LG, Chen W, Schu M, Yaspan BL, Yu Y, Thorleifsson G, Zack DJ, Arakawa S, Cipriani V, Ripke S, et al. Seven new loci associated with age-related macular degeneration
. Nat Genet
2013; 45 (4): 433-9439e1-2, 2013.
73. Huang L, Zhang H, Cheng CY, Wen F, Tam PO, Zhao P, Chen H, Li Z, Chen L, Tai Z, et al. A missense variant in FGD6 confers increased risk of polypoidal choroidal vasculopathy. Nat Genet
48 (6):640–647, 2016.
74. Maguire MG, Ying GS, Jaffe GJ, Toth CA, Daniel E, Grunwald J, Martin DF, Hagstrom SA. Single-nucleotide polymorphisms associated with age-related macular degeneration
and lesion phenotypes in the comparison of age-related macular degeneration treatments trials
. JAMA Ophthalmol
134 (6):674–681, 2016.
75. Pan Y, Chen W, Xu Y, Yi X, Han Y, Yang Q, Li X, Huang L, Johnston SC, Zhao X, et al. Genetic polymorphisms and clopidogrel efficacy for acute ischemic stroke or transient ischemic attack: a systematic review and meta-analysis. Circulation
135 (1):21–33, 2017.
76. Zhao M, Wang X, He M, Qin X, Tang G, Huo Y, Li J, Fu J, Huang X, Cheng X. Homocysteine and stroke risk: modifying effect of methylenetetrahydrofolate reductase C677T polymorphism
and folic acid intervention. Stroke
48 (5):1183–1190, 2017.
77. Lin M, Todaro M, Chan J, Churilov L, Zhu WS, Ramdave S, Mitchell PJ, Dowling RJ, Kwan P, Yan B. Association between CYP2C19 polymorphisms and outcomes in cerebral endovascular therapy. AJNR Am J Neuroradiol
37 (1):108–113, 2016.
78. Muehlschlegel JD, Liu KY, Perry TE, Fox AA, Collard CD, Shernan SK, Body SC. Chromosome 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation
122: (11 suppl): S60–S65, 2010.
79. Gaudino M, Andreotti F, Zamparelli R, Di Castelnuovo A, Nasso G, Burzotta F, Iacoviello L, Donati MB, Schiavello R, Maseri A, et al. The -174G/C interleukin-6 polymorphism
influences postoperative interleukin-6 levels and postoperative atrial fibrillation. Is atrial fibrillation an inflammatory complication? Circulation
2003; 108: Suppl 1:Ii195-9, 2003.
80. Philip I, Plantefeve G, Vuillaumier-Barrot S, Vicaut E, LeMarie C, Henrion D, Poirier O, Levy BI, Desmonts JM, Durand G, et al. G894T polymorphism
in the endothelial nitric oxide synthase gene is associated with an enhanced vascular responsiveness to phenylephrine. Circulation
99 (24):3096–3098, 1999.
81. Lobato RL, White WD, Mathew JP, Newman MF, Smith PK, McCants CB, Alexander JH, Podgoreanu MV. Thrombomodulin gene variants
are associated with increased mortality after coronary artery bypass surgery
in replicated analyses. Circulation
124: (11 suppl): S143–S148, 2011.
82. Allegra S, Fatiguso G, Baietto L, Corcione S, Favata F, Ariaudo A, Pagani N, Ranieri VM, De Rosa FG, Di Perri G, et al. Pharmacogenomic influence on sepsis outcome in critically ill patients. Infez Med
25 (1):45–49, 2017.
83. Nan H, Hutter CM, Lin Y, Jacobs EJ, Ulrich CM, White E, Baron JA, Berndt SI, Brenner H, Butterbach K, et al. Association of aspirin and NSAID
use with risk of colorectal cancer according to genetic variants. JAMA
313 (11):1133–1142, 2015.
84. Diouf B, Crews KR, Lew G, Pei D, Cheng C, Bao J, Zheng JJ, Yang W, Fan Y, Wheeler HE, et al. Association of an inherited genetic variant with vincristine-related peripheral neuropathy in children with acute lymphoblastic leukemia
313 (8):815–823, 2015.
85. Frey UH, Muehlschlegel JD, Ochterbeck C, Fox AA, Shernan SK, Collard CD, Lichtner P, Peters J, Body S. GNAS gene variants
affect beta-blocker-related survival after coronary artery bypass grafting. Anesthesiology
120 (5):1109–1117, 2014.
86. Chung WH, Chang WC, Lee YS, Wu YY, Yang CH, Ho HC, Chen MJ, Lin JY, Hui RC, Ho JC, et al. Genetic variants associated with phenytoin-related severe cutaneous adverse reactions. JAMA
312 (5):525–534, 2014.
87. Lin P, Hartz SM, Zhang Z, Saccone SF, Wang J, Tischfield JA, Edenberg HJ, Kramer JR, M.G A, Bierut LG, et al. A new statistic to evaluate imputation reliability. PLoS One
5 (3):e9697, 2010.
88. Marees AT, de Kluiver H, Stringer S, Vorspan F, Curis E, Marie-Claire C, Derks EM. A tutorial on conducting genome-wide association studies: quality control
and statistical analysis. Int J Methods Psychiatr Res
27 (2):e1608, 2018.