Materials & Methods


1. Systematic Review and Study Selection

The review and selection of studies were carried out between March to May 2020 in the GEO [22] and ArrayExpress [23] public repositories. During this step, the guidelines of the PRISMA declaration for the elaboration of systematic revisions and meta-analysis were followed [20].

The search identified a range of transcriptomics studies related to human AUD. Keywords used during this step included but were not limited to: “transcriptomics,” “alcoholism,” “alcohol abuse,” “alcohol dependence,” “alcohol,” “ethanol,” “Alcohol Use Disorder” and “Homo sapiens.” From this set of studies, those that fulfilled the inclusion criteria were selected, which included data derived from RNA sequencing or genetic expression microarray platforms; the study had information about the sex of the subjects; the study had not been performed on cell lines; the study included a control group; and a minimum size of three subjects per experimental group. The normalized data of selected studies were downloaded using the R package GEOquery [24].


2. Bioinformatics Analysis Strategy

The same strategy was applied to the transcriptomic analysis of each selected study. This analysis included: data preprocessing, differential expression analysis, and functional enrichment analysis. Next, the functional results of all studies were integrated using meta-analysis techniques (Figure 1a depicts the bioinformatics analysis pipeline). Version 3.5.1 of R software [25] was used during the whole study. Computer code is available at https://gitlab.com/ubb-cipf/metafunr.


3. Data Processing and Exploratory Analysis

Data preprocessing included the standardization of the nomenclature of the experimental group of each selected study, focusing on sex and diagnosis of AUD. The probe identifiers from the different platforms were also standardized. The Entrez code of the National Center for Biotechnology Information (NCBI) [26] was used for this step. Repeated probes were summarized using the median of their expression levels. An exploratory data analysis was then carried out (descriptive analysis of the expression levels, principal components analysis, and clustering analysis) to enable the identification of subjects with anomalous behavior or possible batch effects (Figures 1b and c).

4. Differential Expression Analysis and Functional Profiling

The analysis of differential expression levels between sexes was performed by using the R package limma [27]. For every gene, a linear model was adjusted. These models included the contrast to detect differences between women and men when comparing AUD and control groups:
(AUD Women - Control Women) - (AUD Men - Control Men)

P-values associated with the resulting statistics were adjusted using the Benjamini and Hochberg (BH) method [28]. Functional enrichment analysis was performed on the results of the differential expression analysis of each study. This functional profiling was performed using the GSEA method [17], implemented in the R package mdgsa [29]. P-values obtained for every function were corrected again using the BH method. Functions with an adjusted p-value lower than 0.05 were considered statistically significant. The metabolic pathways of the Kyoto Encyclopedia of Genes and Genomes (KEGG) [30,31,32] and the Gene Ontology (GO) [33, 34] were used for this functional enrichment analysis. GO terms were propagated separately for the three ontologies of this database: biological processes (BP), molecular functions (MF), and cellular components (CC).

For each ontology (BP, CC, and MF) and KEGG pathways, we analyzed the number of over-represented elements shared by the studies. These results were graphically represented as UpSet plots [35] to depict the number of elements in common between the different sets.

5. Meta-analysis

Results of the functional characterization of studies were integrated through a meta-analysis, which used the R packages metafor [36] and mdgsa [18]. First, the association with men and women of every KEGG pathway or GO term that appeared in at least two of the analyzed studies was determined. This process was performed using the odds ratio logarithms obtained using the DerSimonian & Laird (DL) method [41] available in the metafor package. This model allowed the detection of functions overrepresented in the set of analyzed studies, with better precision than that offered by the individual analysis previously performed, and thus, offering greater statistical power. In the global estimation of the measured effect, the variability of the individual studies was incorporated, thereby granting greater statistical weight to studies whose values were less variable. The suitability of each analyzed studies was evaluated and confirmed with a heterogeneity study of the aforementioned indicators.

For each one of the KEGG pathways and GO terms analyzed during the meta-analysis, the p-value, the logarithm of the odds ratio (LOR), and its confidence interval were calculated. P-values were adjusted using the BH method, and a particular term was considered significant if it had a p-value lower than 0.05. Significant terms with a LOR greater than 0 indicated an overrepresentation in women, while those with a LOR lower than 0 indicated an overrepresentation in men. Funnel plots and forest plots were used to evaluate the variability and the effect measure of every term in each one of the analyzed studies (Figures 1d and e). The significant results were represented graphically through dot plots and treemaps.

A total of 12,078 BP terms, 1,723 CC terms, 4,182 MF terms, and 229 KEGG pathways were evaluated during the meta-analysis.




Figure 1. (a) Data-analysis workflow. (b) Principal Component Analysis plot. (c) Clustering. (d) A forest plot of the GO:1900391, showing the LOR (log odds ratio) of each study and the global result. (e) Funnel plot of the GO:2001028; dots in the white area indicates the absence of bias and heterogeneity.



References

  1. World Health Organization. Global status report on alcohol and health 2018 (ed. Poznyak, V. and Rekve, D.) ISBN: 978-92-4-156563-9 (World Health Organization 2018).
  2. World Health Organization. International Statistical Classification of Diseases and Related Health Problems, 11th revision. Volume 6. https://icd.who.int/browse11/l-m/en 2018.
  3. Oscar-Berman, M. and Marinković, K. Alcohol: effects on neurobehavioral functions and the brain. Neuropsychol Rev. 17(3), 239-257 (2007).
  4. De Bellis, M. D. et al. Hippocampal volume in adolescent-onset alcohol use disorders. Am J Psychiatry. 157(5), 737-744 (2000).
  5. Olsen, R. W. and Liang, J. Role of GABA A Receptors in Alcohol Use Disorders Suggested by Chronic Intermittent Ethanol (CIE) Rodent Model. Mol Brain. 10(1), 45 (2017).
  6. Koob, G. F. et al. Neurocircuitry targets in ethanol reward and dependence. Alcohol Clin Exp Res. 22(1), 3-9 (1998).
  7. Costardi, J. V. et al. A review on alcohol: from the central action mechanism to chemical dependency. Rev Assoc Med Bras (1992). 61(4), 381-387 (2015).
  8. Pascual, M., Baliño, P., Alfonso-Loeches, S., Aragón, C. M. and Guerri, C. Impact of TLR4 on behavioral and cognitive dysfunctions associated with alcohol-induced neuroinflammatory damage. Brain Behav Immun. 25 Suppl 1, S80-S91 (2011).
  9. Enoch, M. A. Genetic and environmental influences on the development of alcoholism: resilience vs. risk. Ann N Y Acad Sci. 1094, 193-201 (2006).
  10. Wilnsack, S. C., Vogeltanz, N. D., Klassen, A. D. and Harris, T. R. Childhood sexual abuse and women’s substance abuse: national survey findings. J Stud Alcohol. 58(3), 264-271 (1997).
  11. Edenberg, H. J. The genetics of alcohol metabolism: role of alcohol dehydrogenase and aldehyde dehydrogenase variants. Alcohol Res Health. 30(1), 5-13 (2007).
  12. Chen, C. H., Ferreira, J. C., Gross, E. R. and Mochly-Rosen, D. Targeting aldehyde dehydrogenase 2: new therapeutic opportunities. Physiological Reviews. 94(1), 1-34 (2014).
  13. Neumark, Y. D., Friedlander, Y., Thomasson, H. R. and Li, T. K. Association of the ADH2*2 allele with reduced ethanol consumption in Jewish men in Israel: a pilot study. J Stud Alcohol. 59(2), 33-39 (1998).
  14. Witt, E. D. Puberty, hormones, and sex differences in alcohol abuse and dependence. Neurotoxicology and Teratology. 29(1), 81-95 (2007).
  15. Garfinkel, L., Boffetta, P. and Stellman, S. D. Alcohol and breast cancer: a cohort study. Prev Med 17, 686-693 (1988).
  16. Manzoni, C. et al. Genome, transcriptome and proteome: The rise of omics data and their integration in biomedical sciences. Brief Bioinform. 19(2), 286-302 (2018).
  17. Subramanian, A. et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences. 102, 15545-15550 (2005).
  18. García-García, F. Methods of functional enrichment analysis in genomic studies. Ph.D. thesis. (2016).
  19. Normand, S. L. T. Meta-analysis: formulating, evaluating, combining, and reporting. Statistics in Medicine. 18, 248-287 (1999).
  20. Liberati, A. et al. The prisma statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. PLOS Medicine. 6, 1-28 (2009).
  21. Voight, A. P. et al. Single-cell transcriptomics of the human retinal pigment epithelium and choroid in health and macular degeneration. Proc Natl Acad Sci U S A. 116(48), 24100-24107 (2019).
  22. Barrett, T. et al. NCBI GEO: archive for functional genomics data sets—update. Nucleic Acids Research. 41, D991-D995 (2012).
  23. Kolesnikov N. et al. 2015 ArrayExpress update - simplifying data submissions. Nucleic Acids Res. 43(Database issue), D1113-1116 (2015).
  24. Davis, S. and Meltzer, P. GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics. 14, 1846-1847 (2007).
  25. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2019).
  26. Maglott, D., Ostell, J., Pruitt, K. D. and Tatusova, T. Entrez Gene: gene-centered information at NCBI. Nucleic acids research. 39, D52-D57 (2010).
  27. Ritchie, M. E., et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015).
  28. Benjamini, Y. and Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B (Methodological). 57, 289-300 (1995).
  29. Montaner, D. and Dopazo, J. Multidimensional gene set analysis of genomic data. PLOS ONE. 5, 1-13 (2010).
  30. Kanehisa, M., Sato, Y., Furumichi, M., Morishima, K., and Tanabe, M. New approach for understanding genome variations in KEGG. Nucleic Acids Res. 47, D590-D595 (2019).
  31. Kanehisa, M., Furumichi, M., Tanabe, M., Sato, Y., and Morishima, K. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 45, D353-D361 (2017).
  32. Kanehisa, M. and Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 28, 27-30 (2000).
  33. Ashburner, M., et al. Gene ontology: tool for the unification of biology. Nat Genet. 25(1), 25-29 (2000).
  34. The Gene Ontology Consortium. The Gene Ontology Resource: 20 years and still GOing strong. Nucleic Acids Res.47(D1), D330-D338 (2019).
  35. Lex, A. and Gehlenborg, N. Sets and intersections. Nat methods. 11, 779 (2014).
  36. Viechtbauer, W. Conducting Meta-Analyses in R with the metafor Package. Journal of Statistical Software. 36, 1-48 (2010).
  37. McClintick, J. et al. Stress-response pathways are altered in the hippocampus of chronic alcoholics. Alcohol. 47(7), 505-515 (2013).
  38. Xu, H. et al. Sex-biased methylome and transcriptome in human prefrontal cortex. Human Molecular Genetics. 23(5), 1260-1270 (2014).
  39. McClintick, J. et al. Ethanol treatment of lymphoblastoid cell lines from alcoholics and non-alcoholics causes many subtle changes in gene expression. Alcohol. 48(6), 603-610 (2014).
  40. Beech, R. et al. Stress-related alcohol consumption in heavy drinkers correlates with expression of miR-10a, miR-21, and components of the TAR-RNA-binding protein-associated complex. Alcoholism, clinical and experimental research. 38(11), 2743-2753 (2014).
  41. DerSimonian, R. and Laird, N. Meta-analysis in clinical trials. Controlled Clinical Trials. 7, 177-188 (1986).
  42. Kim, A. M., Tingen, C. M. and Woodruff, T. K. Sex bias in trials and treatment must end. Nature. 465(7299), 688-689 (2010).
  43. Wilkinson, M. D. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data. 3, 160018 (2016).
  44. European Commission. Guidelines on FAIR Data Management in Horizon 2020. Tech. Rep. (2016).
  45. de la Monte, S. M. and Kril, J. J. Human alcohol-related neuropathology. Acta Neuropathol. 127(1), 71-90 (2014).
  46. Pfefferbaum, A., Rosenbloom, M., Rohlfing, T. and Sullivan, E. V. Degradation of association and projection white matter systems in alcoholism detected with quantitative fiber tracking. Biol Psychiatry. 65, 680-690 (2009).
  47. Crews, F. T. and Nixon, K. Mechanisms of neurodegeneration and regeneration in alcoholism. Alcohol Alcohol. 44(2), 115-127 (2009).
  48. Becker, J., McClellan, M. and Reed B. G. Sex differences, gender and addiction. J Neurosci Res. 95(1-2), 136-147 (2017).
  49. Xie, Q., Buck, L., Bryant, K. and Barker, J. Sex Differences in Ethanol Reward Seeking Under Conflict in Mice. Alcohol Clin Exp Res. 43(7), 1556-1566 (2019).
  50. Romanazzi, V., Schilirò, T., Carraro, E. and Gilli, G. Immune response to acetaldehyde-human serum albumin adduct among healthy subjects related to alcohol intake. Environmental Toxicology and Pharmacology. 36(2), 378-383 (2013).
  51. Petrasek, J. et al. Metabolic danger signals, uric acid and ATP, mediate inflammatory cross-talk between hepatocytes and immune cells in alcoholic liver disease. J Leukoc Biol. 98(2), 249-256 (2015).
  52. De Santis, S. et al. Chronic alcohol consumption alters extracellular space geometry and transmitter diffusion in the brain. Science Advances. 6(26), eaba0154 (2020).
  53. Pavarin, R. M. et al. Mortality risk among individuals treated for alcohol use disorders: results of a longitudinal study from 1978 to 2016 in Northern Italy. Eur Rev Med Pharmacol Sci. 24(4), 1995-2005 (2020).
  54. Dennis, M. K. et al. BLOC-1 and BLOC-3 regulate VAMP7 cycling to and from melanosomes via distinct tubular transport carriers. J Cell Biol. 214(3), 293-308 (2016).
  55. Chanaday, N. L., Cousin, M. A., Milosevic, I., Watanabe, S. and Morgan, J. R. The Synaptic Vesicle Cycle Revisited: New Insights into the Modes and Mechanisms. J Neurosci. 39(42), 8209-8216 (2019).
  56. Mooney, S. M. and Miller, M. W. Time-specific effects of ethanol exposure on cranial nerve nuclei: gastrulation and neuronogenesis. Exp Neurol. 205(1), 56-63 (2007).
  57. Guerri, C. and Pascual, M. Reproductive and developmental toxicology. Second Edition.. (ed. Gupta, R. C.) Chapter 24: Effects of Alcohol on Embryo/Fetal Development. (Academic Press 2017) ISBN: 9780123820327 .
  58. Schaffer, A. E., Pinkard, O. and Coller, J. M. tRNA Metabolism and Neurodevelopmental Disorders. Annu Rev Genomics Hum Genet. 20, 359-387 (2019).