Polygenic influences on the behavioral effects of alcohol withdrawal in a mixed-ancestry population from the collaborative study on the genetics of alcoholism (COGA).

Document Type


Publication Date



JMG, Male, Female, Humans, Adult, Alcoholism, Substance Withdrawal Syndrome, Genome-Wide Association Study, Alcohol Drinking, Phenotype, Polymorphism, Single Nucleotide

JAX Source

Mol Cell Neurosci. 2023;125:103851








The following body of work was supported by a grant from the Na- tional Institute on Drug Abuse (DP1DA042103 awarded to RHCP and R01DA048890 awarded to JB). CEBB is also supported by R01DA042742. RHCP is also supported by R01DA037927. Data for current study was provided by the Collaborative Study on the Genetics of Alcoholism which was supported by NIH Grant U10AA008401. None of these organizations had any role in study conception, design, or interpretation, or the decision to publish these data. The findings and conclusions in this publication are those of the authors, and do not represent the views of the U.S. Department of Veterans Affairs, the U.S. Department of Agriculture, and do not represent any US Government determination, position, or policy.


Alcohol withdrawal (AW) is a feature of alcohol use disorder that may occur in up to half of individuals with chronic, heavy alcohol consumption whenever alcohol use is abruptly stopped or significantly reduced. To date, few genes have been robustly associated with AW; this may be partly due to most studies defining AW as a binary construct despite the multiple symptoms and their range in severity from mild to severe. The current study examined the effects of genome-wide loci on a factor score for AW in high risk and community family samples in the Collaborative Study for the Genetics of Alcoholism (COGA). In addition, we tested whether differentially expressed genes associated with alcohol withdrawal in model organisms are enriched in human genome-wide association study (GWAS) effects. Analyses employed roughly equal numbers of males and females (mean age 35, standard deviation = 15; total N = 8009) and included individuals from multiple ancestral backgrounds. Genomic data were imputed to the HRC reference panel and underwent strict quality control procedures using Plink2. Analyses controlled for age, sex, and population stratification effects using ancestral principal components. We found support that AW is a polygenic disease (SNP-heritability = 0.08 [95 % CI = 0.01, 0.15; pedigree-based heritability = 0.12 [0.08,0.16]. We identified five single nucleotide variants that met genomewide significance, some of which have previously been associated with alcohol phenotypes. Gene-level analyses suggest a role for COL19A1 in AW; H-MAGMA analyses implicated 12 genes associated with AW. Cross-species enrichment analyses indicated that variation within genes identified in model organism studies explained <1 >% of the phenotypic variability in human AW. Notably, the surrounding regulatory regions of model organism genes explained more variance than expected by chance, indicating that these regulatory regions and gene sets may be important for human AW. Lastly, when comparing the overlap in genes identified from the human GWAS and H-MAGMA analyses with the genes identified from the animal studies, there was modest overlap, indicating some convergence between the methods and organisms.