Jeff K – Multi-Omics; Data Integration in Addiction
It has long been recognized that substance use disorders are highly complex and exist along a spectrum of disease severity. Predisposing factors include genetics, individual environmental factors, family dynamics, social and cultural differences, and co morbid conditions.
This post begins with an overview of select factors which may be considered in an integrated approach. This is followed by examples of recent studies along with explanation of some of the techniques employed.
At the neurobiological level while common pathways have been identified at the macroscopic level, neural systems and subsystems at the cellular level are still incompletely understood.
Data integration studies at the cellular level include regulation and signaling of gene expression, epigenetics, and proteinomics. Coordination and networking between brain regions and functional anatomy is an area of active investigation.
There is an array of therapeutic approaches available for treatment of substance use disorders. A significant fraction estimated in the 50-70% range recover with little or no outside intervention. Clinical psychotherapy includes residential or outpatient treatment, and individual therapy encompassing a number of techniques varying from one center to another. There are a number of peer support groups available to individuals including SMART, LifeRing, recovery dharma and 12 step groups all of which now have online virtual meetings increasing choice and availability.
There are three FDA approved medications for treatment of alcohol use disorder, naltrexone, acamprosate and disulfiram and at least three others used off label. The GLP-1 agents in preliminary studies appear highly effective. Medications and choice of psychosocial modalities is often by trial and error.
Notably project MATCH a large multi center eight year study sponsored by the NIAAA launched in 2008 with the goal of finding patient characteristics capable of optimally assigning individuals to one of three proven treatment modalities failed in that goal.
The goal of an individualized approach to SUD treatment and prevention may be closer due to advances in neuroscience and data processing. This post explores the emerging science of multi-omics which integrates biological, psychosocial, and demographic data with the aim of developing iterative diagnostic, prognostic, and therapeutic approaches tailored to the needs of the individual.
A table of acronyms used and definitions is included toword the end of this post. Select references illustrating techniques referred to are included in green backgrounds. References used in preparation of this post with links are included at the end of the post.
This graphic illustrates fields with potential for developing a comprehensive approach to treatment for complex disease states such as substance use disorders.
Many early studies have been lacking in reproducibility and specificity due to insufficient case numbers and lack of diversity in subject populations. As larger open databases and more cost and time efficient techniques become available these difficulties are beginning to be overcome.
Advances in data processing AI and machine learning have the potential to identify unrecognized disease mechanisms through identification and correlations without pre existing bias and hypothesis.
Substance use disorders have a strong heritability component. Twin studies estimate heritability factor of approximately 40-60%. Principles of genetics have been covered in a previous post. The studies we will be looking at in this post are largely based in genetic correlations.
The candidate gene approach involves selection of a gene involved in a known mechanism, dopamine receptors for example and then initiating population and functional studies relating to the disease process. This has yielded limited information and has largely been abandoned. The multi genetic diverse nature of addiction does not lend itself to hypothesis driven approaches.
Genome wide association studies are capable of screening many thousands of samples with some studies encompassing a million or more individuals resulting in greater statistical power. The most common method is to screen for single base substitutions known as single nucleotide polymorphisms – SNPs. An SNP occurs when there is a single base substitution at one location. There are over 5 million known SNPs.
It is also possible to screen for exomes (WES). Exomes are segments of DNA that are transcribed into RNA as opposed to non coding DNA. Up to 85% of genes involved in the disease process may be identified by this method.
Whole genome sequencing (WGS) is also possible. These techniques are more time consuming and costly limiting sample size. WGS may result in inconclusive results due to inherent variation and rare variants.
In a polygenetic condition such as alcohol use disorder or coronary artery disease, genetic load itself is not sufficient to produce the disease (phenotype). Rather, genes can influence how an organism will respond to an environmental stimulus. The above graph illustrates the relative contributions of family environment, genetics, and external environmental factors in SUD.
At early ages through late adolescence family factors have the largest influence. Through the later teens through adulthood family takes a decreasing role and genetic factors along with the external environment increase in importance.
This diagram breaks down the process by which an activated gene results in a protein.
– Transcription is the creation of an RNA copy of a gene.
– Splicing of the initial copy cuts out unused introns creating messenger mRNA composed of exons which leaves the nucleus.
– Variations in splicing make it possible to create differering protein variations from a single gene.
– mRNA attaches to a ribosome and an amino acid chain corresponding to the coded sequence is assembled into a final folded protein.
– Even a small change in amino acid sequence and final shape of the protein can significantly alter function.
– DNA is a code for protein synthesis
Each of these steps is a potential target for study and pharmacological treatment. AI is now used in research to model new agents matching designer drugs to receptors for example. This can greatly speed up drug design compared to trial and error methods. It may also allow for targeting at the individual level.
EEG and imaging studies are promising areas to investigate in multiomics studies. They are objective, reproducible methods to localize and characterize brain activity. EEG by electrophysiology and functional fMRI indirectly measuring focal brain activity by hemodynamics.
Both methods are capable of localizing brain activity at rest or in response to a stimulus or task. They are non invasive proven technologies. Cost and time constraints limit size of study populations.
Demographic and individual factors including family history, age, disease severity, substance(s) used and comorbid contritions are ,important co factors to consider in analysis of any data collected,
Rather than seeing environmental and personal factors as confounding variables such factors may be incorporated into predictive markers and individual treatment
The studies in these examples begin with single nucleotide polymorphism (SNP) genome wide associations (GWAS) correlated to a phenotype such as alcohol use disorder (AUS).
The entire genome can be screened for single base substitutions.
The frequency of occurrence of individual SNPs is represented by the colored bell shaped distribution curves.
The lower graph is named a manhattan plot for its resemblance to a skyline. Each color represents a different chromosome. Spikes reflect log P values. Because most variants occur by chance a very high threshold is required to isolate SNPs occurring almost exclusively in the trait study group.
In the human genome of 3 billion base pairs many variations are present unrelated to the trait under study. The significance threshold is commonly set at 0.00000001% probability of occurring by chance.
Technical notes are in green background.
Microarray is a technique allowing for high volume analysis of DNA. The process begins with extraction of DNA from host tissue such as blood or cheek swab. Following amplification the DNA is fragmented and a library of segments is prepared and tagged with a fluorescent marker. The reference sample is also tagged with a different color marker
The microarray chip is composed of wells each containing a mirror DNA sequence each embedded with an SNP or segment composed of the complementary base sequence. If the subject or reference sample matches it will bind to the well and the resulting fluorescent color can be detected and read out by automated device.
Results can then be analyzed and by correlating with a large volume of samples represented as a table or graph with a probability distribution.
note: this a simplified working explanation of the statistical test. Please refer to reference texts for more information
P values are a common method of determining the statistical significance of data when comparing a test result with a reference standard. In this case the distribution of a DNA polymorphism in a population meeting criteria for AUD compared with the general population. Under the chosen test model it reflects probability relative to the null hypothesis with a value between 0 and 1. The null hypothesis here would be that the result is due to chance.
So smaller p-values reflect greater probability that the polymorphism is not due to chance and is due to an effect size such as association with AUD. Note that it does not prove that the test hypothesis is true and does not reflect the strength of association.
GAWS of a population of 2697 nicotine dependent individuals. Subjects were divided by ancestry as European American or African American. Previous studies have been limited by over representation of European descent individuals and one aim of the study was to correct for this imbalance. Populations were drawn from the collaborative Study of Addiction Genetics and Environment project.
Primary aims of the study were to correlate GWAS findings with nicotine dependence (ND) severity score and cellular mechanisms related to nicotine dependance.
Above manhattan plots reflect SNPs identified in the study populations. There were two significant SNPs found in the European American population and eight in the African American sample.
This is a closeup map of a segment cluster associated with ND on chromosome 8. Color coding reflects associated ND dependence severity score, This demonstrates the potential for developing bio markers reflecting quantitative disease severity rather than a more limited binomial test.
This is a correlation between GWAS findings and a cellular biological pathway. The above diagram reflects the endothelial nitric acid sythase pathway correlated to SNPs associated with ND in the studied population.
This pathway has an essential role in repair and maintenance of blood vessel walls and is associated with hypertension. The purple highlighted proteins have polymorphisms implicated in substance use disorders.
Technical note: the following study utilizes EEG data.
Electroencephalograms are a well established diagnostic tool in clinical medicine and in research. The study uses scalp electrodes placed most often in the above 10-20 system. Standardized placement allows for comparisons and longitudinal data.
The map above shows color coded lobar location of electrode placement.
Recorded activity reflects local electrical currents arising from cortical brain regions corresponding to electrode locations. Currents are generated by synaptic ion flow, not action potentials. The waveforms are characterized by frequency grouped as described in the above table.
The lower to higher frequency ranges indicate states of relative lower to higher neural activity during sleep or while awake. The test is non invasive and data can be collected and correlated in longitudinal studies.
This study recruited 2832 individuals of African ancestry enrolled in the COGA longitudinal study with the aim of developing an objective bio marker for development of AUD based on genome SNP identification combined with EEG findings, specifically fast β EEG.
African ancestry was chosen due to over representation of European ancestry in previous studies limiting significance of findings often not reproducible in broader population bases.
The fast β EEG pattern has been noted to correlate with other externalizing disorders such as ADHD and internet addiction suggesting a common mechanism. It has also been shown to be predictive of relapse in abstinent alcohol dependent individuals.
This plot reflects GWAS correlating to both AUD and presence of the fast β EEG pattern, A single locus correlated to both nicotine dependance and EEG findings was isolated on chromosome 3 position 3q26.
Detailed close up chart of GWAS at location of the identified SNP rs12729469. The SNP lies in a stretch if non coding DNA adjacent to several genes. Non coding DNA may serve a role in gene expression and may result in increased or decreased gene expression.
Expression Quantitative trait localization (eQTL) analysis was performed at this location. The method analyzes gene expression by measuring RNA production at associated genes. Decreased expression of the gene coding for the protein pseudocholinesterase (BCHE) located just upstream of the SNP identified. eQTL allows for quantitative correlation of cellular activity with expressed traits such as memory or behavior.
BCHE is involved in breakdown of the neurotransmitter acetylcholinesterase in the thalamus. This area in important in memory and cognition. The polymorphism then plausibly has a role in development of nicotine dependence. The combination of fast β EEG and presence of SNP rs12729469 could potentially serve as one biomarker for nicotine use disorder.
The next study involves development of a multi factorial polygenetic risk score predictive of remission from alcohol use disorder. The study involved 1376 individuals from the COGA longitudinal study population. Multivariate analysis was used with a machine learning approach.
The first stage is to assemble a GWAS of the case and control populations for the phenotypes included in the analysis. This may include AUD remission, impulsivity, co-existing disorders, or other factors.
A weight for each allele can then be determined for each SNP correlated to the outcome measure. Predictive risk score is then the normalized sum of the weighted factors.
Note: this is a brief summary of Support Vector Learning please refer to reference sources for more complete explaination
Support Vector Machine Learning is a type of machine learning often used to separate data into a binary +/- classification.
The method creates a plane with maximal distance between the two closest data points. These are referred to as the support vectors. These are circled in red in the example above.
SVM is an efficient way to separate data into two sets such as normal vs abnormal and there are adaptations that can be used for more complex or non linear data sets. It is efficient in memory use as only the support vectors are required for calculation.
This study involved 1376 subjects enrolled in COGA all of whom met criteria for active AUD at onset. The group was followed over time in order to develop a predictive index for remission. Subjects were grouped by European American or African American ancestry and divided by sex.
The above table lists demographic data with calculated weights for each factor predictive of remission.
The predictive risk score was derived from genetic GWAS, demographic variables such as employment, marital status, or medications along with EEG data. These factors were used to construct a support vector machine algorithm predictive of remission from AUD.
The graph illustrates accuracy for select combinations of these elements with some in the 85% range. If further developed and verified by additional studies these could reach accuracy sufficient for clinical utility. There are currently no such verified measures used to guide treatment decisions and identify individuals requiring additional or more intensive therapy.
The study also utilized EEG data to create group specific connectivity maps of brain regions demonstrating increased or decreased connection predictive of AUD remission.
Circled in red are default mode structures. Changes in default mode connections were prominent in observed results.
Note: a polymorphism does not need to be located in a coding segment of DNA to be functionally significant.
97% of DNA in the human genome is non-coding, meaning it does not directly code for a protein. Formerly described as “junk DNA” it is now known that non coding DNA often is important in gene expression, regulation, and structure.
Transcription, epigenetics, and factors regulating gene expression are active areas of research and can be targets in development of more specific guided pharmacotherapy.
Up to this point we have looked at single base substitution SNP studies as baselines then correlated with additional inputs such as EEG or demographics to develop models. SNP based analysis is the simplest, most cost and time efficient method to screen the large number of samples required for functional genetic studies.
A large gap exists between the 50% heritability factor for SUDs derived from twin studies and relatively small number accounted for by GWAS. A large portion of this gap is due to the very high significance thresholds inherent in whole genome studies. To locate variations with explanatory power it is necessary to explore transcription, epigenetics, RNA splicing variation, and protein synthesis.
Gene expression can be thought of as networks where activation or suppression of one gene results in suppression or enhancement of one or more nearby or distant related genes.
Multivariate association studies aim to correlate gene expression and other related variables to arrive at correlations useful in clinical contexts.
The largest GWAS to date combined data from over one million individuals correlated to risk of general substance use disorder. This study utilized this data set to derive a composite addiction risk score addiction-rf.
Based on addiction-rf transcription wide association study (TWAS) was conducted utilizing mRNA data localized to specific brain structures from donated cadaveric brain tissue. The plots above identify specific genes actively transcribed associated with clinical SUD. Color coding identifies specific structural brain locations. The two graphs reflect data derived from different databanks.
The risk score addiction-rf was further correlated to a data bank of 1,147 traits to derive a phenotype wide association study. This included information such as employment status, maternal smoking at birth, and history of ADHD. Points above the baseline represent positive effect and below represent a negative direction of effect.
Predictive value of the addiction-rf for individual substances. The highest values were for two or more instances of poly substance and tobacco use disorders. The lowest was for cannabis use disorder. The values are relative risks and do not reflect diagnostic sensitivity or specificity.
Developing an individualized approach for the diagnosis and treatment of a complex neuropsychiatric disorder like SUD is challenging. However the high relapse rates and the array of treatment choices each of which alone has a relatively low success rate suggests that a more tailored approach is needed.
Improvements in data processing, use of AI, and machine learning allow for studies using larger population sizes. Large open access databases are now available. Hypothesis free study designs are likely to find unexpected correlations leading to discovery of new mechanisms and therapeutic targets.
Psychiatry and SUD have lagged behind in investment and development of new medications. Most often promising medications derived from animal experiments are ineffective in human trials or are poorly tolerated.
Precision pharmacotherapy has been best developed in medical oncology where precise identification of tumor markers allows for improved targeted drug therapy. Each case has a multidisciplary approach with close monitoring of therapeutic response.
In addiction treatment there has traditionally been a “one size fits all” approach, available medications are underprescribed, and professional therapy is non standardized.
Despite these challenges significant advances leveraging existing technologies has the potential to improve patient choice and therapeutic efficacy.
The NIH is the largest single funding source for research into SUD in the world and the US is the global leader in the field and in training of new research scientists. The NIDA has a current annual budget of $1.6 billion. The economic cost in the US attributed to alcohol and drug abuse in $820 billion per year.
This post reviewed some of the technical advances in the emerging field of multi-omics which aims to integrate data across multiple fields to better understand complex processes involved in substance use disorders.
Using improved data processing and analytics biological, psychosocial, and clinical data can be integrated to produce better understanding of interrelated processes and useful disease markers.
A common goal in addiction treatment and general medicine is to move to a more personal individually tailored approach to optimize diagnosis and treatment selection and relapse risk, and individually tailored pharmacotherapy. This effort is in the early stages of development.
Acronyms and definitions
AUD – alcohol use disorder by DSM criteria
Epigenetics – the field of study involving biological changes affecting how genes work turning them on or off, without changing DNA itself. Often related to environment or development.
Exons – segments of initial gene copy included in final mRNA after splicing
GWAS – Genome Wide Association Study, research study correlating genetic variants (usually SNPs) in a population with a phenotype or trait
Intron – segments of the initial gene copy spliced out during final mRNA assembly
Machine Learning – a subset of artificial intelligence AI, allows computers to learn and accurately classify data over time
ND – nicotine dependance
PRS – polygenetic risk score, composite risk for a trait based on multiple genetic variants
SNP – Single nucleotide polymorphism, a genetic variant consisting of a single base substitution
SUD – Substance use disorder by DSM criteria
SVL – Support vector learning, A type of machine learning used to separate data into two or more classes or groups. Calculates a line or plane of separation with maximal distance from the two closest points termed support vectors
WES – Whole exome sequencing – an association study correlating expressed RNA with a trait or phenotype
WGS – whole genotype sequencing., laboratory analysis of the entire genome, 3 billion base pairs in humans
Thank you for your interest in reading this post. Feedback is always welcome. Jeffk072361@gmail.com
For education and information purposes only. No commercial or interest. This post should not be considered medical or professional advice. Images and data obtained from sources freely available on the World Wide Web.
References
Addiction Science: Uncovering Neurobiological Complexity
N. D. Volkow and R. D. Baler
National Institute on Drug Abuse, National Institutes of Health, Bethesda, MD 2089
Neuropharmacology. 2014 January ; 76(0 0): . doi:10.1016/j.neuropharm.2013.05.007.
…………………………………………………………………..
The addiction risk factor: A unitary genetic vulnerability characterizes substance use disorders and their associations with common correlates
Alexander S. Hatoum1 ✉, Emma C. Johnson 1, Sarah M. C. Colbert
Neuropsychopharmacology (2022) 47:1739–1745; https://doi.org/10.1038/s41386-021-01209-w
The Relationship between the Brain-Derived Neurotrophic Factor Gene Polymorphism (Val66Met) and Substance Use Disorder and Relapse
Aleksandra Strońska-Pluta, Aleksandra Suchanecka, Krzysztof Chmielowiec Jolanta
Int. J. Mol. Sci.2024, 25(2), 788;https://doi.org/10.3390/ijms25020788
Genetics of substance use disorders in the era of big data
Joel Gelernter, Renato Polimanti
Department of Psychiatry, Yale University School of Medicine, West Haven, CT, USA.
Nat Rev Genet . 2021 November ; 22(11): 712–729. doi:10.1038/s41576-021-00377-1.
………………………………………………………………..
GWAS 1 million individuals
Nat Ment Health . 2023 March ; 1(3): 210–223. doi:10.1038/s44220-023-00034-y.
Recent advances in the genetic epidemiology and molecular
genetics of substance use disorders
Kenneth S Kendler, Xiangning Chen, Danielle Dick, Hermine Maes
Nat Neurosci . ; 15(2): 181–189. doi:10.1038/nn.3018.
…………………………………………………………….
Cao, J., Hudziak, J. & Li, D. Multi-Cultural Association of the Serotonin Transporter Gene (SLC6A4) with Substance Use Disorder. Neuropsychopharmacol 38, 1737–1747 (2013). https://doi.org/10.1038/npp.2013.73
https://www.nature.com/articles/npp201373
Homer Isoforms Differentially Regulate Cocaine-Induced
Neuroplasticity
Neuropsychopharmacology (2006) 31, 768–777
Wittea2012EffectsOC,
tEffects of COMT polymorphisms on brain function and behavior in health and disease, Veronica Wittea and Agnes, Brain Research Bulletin, 2012, 88p 418-428 https://api.semanticscholar.org/CorpusID:5338848
……………………………………………………………………..
.The scientific article also stresses the importance of this study in its demonstration that genetic factors may not directly “cause” disease, but instead moderate responses to environmental factors (Caspi, et. al. 2003). This is another important subtlety that is often missed by the popular press.
https://www.bio.davidson.edu/courses/genomics/2003/mccord/5-htt.html
An Evaluation of µ-Opioid Receptor (OPRM1)
as a Predictor of Naltrexone Response
in the Treatment of Alcohol Dependence
Results From the Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence (COMBINE) Study
Raymond F. Anton, MD; Gabor Oroszi, MD, PhD; Stephanie O’Malley, PhD; David Couper, PhD;
Robert Swift, MD, PhD; Helen Pettinati, PhD; David Goldman,
……………………………………………………………………
STASH, Vol. 9(1) – Not so different after all? Sex-specific patterns of psychiatric comorbidity among men and women with SUD
Posted on January 2, 2013
How genome-wide association studies (GWAS) made traditional candidate gene studies obsolete
Laramie E. Duncan, Michael Ostacher & Jacob Ballon
Neuropsychopharmacology volume 44, pages 1518–1523 (2019)
https://www.nature.com/articles/s41386-019-0389-5
The Serotonin Transporter Gene and Risk for Alcohol
Dependence: A Meta-Analytic Review
R. Kathryn McHugh, Stefan G. Hofmann, Anu Asnaani, Alice T. Sawyer, and Michael W. Otto
Drug Alcohol Depend. 2010 April 1; 108(1-2): 1–6. doi:10.1016/j.drugalcdep.2009.11.017.
………………………………………………………………………….
Association between Val66Met Brain-Derived Neurotrophic Factor
(BDNF) Gene Polymorphism and Post-Treatment Relapse in
Alcohol Dependence
Marcin Wojnar, M.D., Ph.D.1,2
, Kirk J. Brower, M.D.
Alcohol Clin Exp Res. 2009 April ; 33(4): 693–702. doi:10.1111/j.1530-0277.2008.00886.x.
………………………………………………………………………………..
New Study Psychiatric Disorders Genetics
https://www.sciencedaily.com/releases/2025/12/251223084855.htm
https://www.nature.com/articles/d41586-025-04037-w
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Unraveling the Complexity of Gene-Environment Interactions in Noncommunicable Diseases through Multi-Omics Research
…………………………………………………………………….
………………………………………………………….
Genetics of substance use disorders in the era of big data
Joel Gelernter Renato Polimanti
Department of Psychiatry, Yale University School of Medicine, West Haven, CT, USA.
Nat Rev Genet . 2021 November ; 22(11): 712–729. doi:10.1038/s41576-021-00377-1.
Rare and common variants associated with alcohol consumption identify a conserved molecular network
Brittany S. Leger, John J. Meredith, Trey Ideker, Sandra Sanchez-Roige, Abraham A. Palmer
………………………………………………….m
Data related to Genetic diversity fuels gene discovery for tobacco and alcohol use
Saunders, Gretchen R B Wang, Xingyan Chen, Fang Jang, Seon-Kyeong Liu, Mengzhen Wang, Chen Liu, Dajiang J Vrieze, Scott
https://conservancy.umn.edu/browse/author?startsWith=Wang,%20Chen
……………………………………………………
Nat Protoc . 2023 June ; 18(6): 1745–1759. doi:10.1038/s41596-022-00797-1.
Mapping the common gene networks that underlie related diseases
Sara Brin Rosenthal, Sarah N. Wright, Sophie Liu, Christopher Churas, Chilin-Fuentes, Chi-Hua Chen, Kathleen M. Fisch, Dexter Pratt, Jason F. Kreisberg, Trey Ideker*,2,3
1Center for Computational Biology & Bioinformatics, Department of Medicine, University of
California, San Diego, La Jolla, California, USA
………………………………………………………
Epigenetic signatures of starting and stopping smoking
Daniel L. McCartney a, Anna J. Stevenson ,
https://www.sciencedirect.com/science/article/pii/S2352396418304717
D.L. McCartney et al. / EBioMedicine 37 (2018) 214–220
………………………………………………………….
Association of OPRM1 Functional Coding Variant With Opioid Use Disorder
A Genome-Wide Association Study
Hang Zhou, PhD, ChristopherT. Rentsch, PhD
JAMA Psychiatry Published Online: June 3, 2020 2020;77;(10):1072-1080. doi:10.1001/jamapsychiatry.2020.1206
https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2766708
………………………………………………………….
https://www.cdc.gov/pulsenet/php/wgs/index.html
Whole Genome Sequencing
…………………………………………………………….
…………………………………………………………………..
https://article.innovationforever.com/ID/20240153.html
Omics, sequencing
……………………………………………………..
You tube genetics relapse
…………………………………………
A Developmental Twin Study of Peer-Group Deviance
1. Kenneth S. Kendler, MD; Kristen C. Jacobson, PhD; Charles O. Gardner, PhD et al
JAMA Psychiatry. 2025; 83(1):51-58. 10.1001/jamapsychiatry.2025.3565
Application of omics-based biomarkers in substance use disorders
Longtao Yang, Lijie Zhang , Huiting Zhang
https://doi.org/10.1016/j.metrad.2023.100008m
Available online 19 June 2023
https://www.sciencedirect.com/science/article/pii/S2950162823000085
………………………………………………………
Quote
Substance use disorder (SUD) is a type of addictive encephalopathy resulting from drug abuse, which leads to
abnormal cerebral alterations indicating neurotoxicity that is manifested through various biomarkers.
……………………………………………………..
Gu et al. Journal of Biomedical Semantics https://doi.org/10.1186/s13326-021-00256-y
(2021) 12:21
Prefrontal fNIRS-based clinical data analysis
of brain functions in individuals abusing
different types of drugs
https://link.springer.com/article/10.1186/s13326-021-00256-y
…………………………………………..
Predicting risk for Alcohol Use Disorder using longitudinal data
with multimodal biomarkers and family history: A machine learning study
Sivan Kinreich, Jacquelyn L. Meyers, Adi Maron-Katz
Mol Psychiatry . 2021 April ; 26(4): 1133–1141. doi:10.1038/s41380-019-0534-x.
Genome-wide association study identifies loci associated with liability to alcohol and drug dependence that is associated with variability in reward-related ventral striatum activity in African- and European-Americans
Leah Wetherill, Dongbing Lai, Emma C. Johnson,
First published: 17 May 2019 https://doi.org/10.1111/gbb.12580 Digital Object Identifier (DOI)
https://onlinelibrary.wiley.com/doi/full/10.1111/gbb.12580
The collaborative study on the genetics of alcoholism: Genetics
Emma C. Johnson, Jessica E. Salvatore,
https://onlinelibrary.wiley.com/doi/full/10.1111/gbb.12580
Support vector machine
………………………………………………………….
The Collaborative Study on the Genetics of Alcoholism: Overview
Arpana Agrawal, Sarah J. Brislin, Kathleen K. Bucholz
September 2023 https://doi.org/10.1111/gbb.12864
https://onlinelibrary.wiley.com/doi/10.1111/gbb.12864
https://onlinelibrary.wiley.com/doi/10.1111/gbb.12864
Predicting alcohol use disorder remission: a longitudinal multimodal multi-featured machine learning approach
Sivan Kinreich 1, Vivia V. McCutcheon2, Fazil Ali
Kinreich et al. Translational Psychiatry (2021) 11:166 https://doi.org/10.1038 s41398-021-01281-2
https://www.nature.com/articles s41398-021-01281-2.pdf
Genome-Wide Association Study of Nicotine Dependence in American Populations: Identification of Novel Risk Loci in Both African-Americans and European-Americans
Joel Gelernter, Henry R. Kranzle
Biol Psychiatry. 2015 March 1; 77(5): 493–503. doi:10.1016/j.biopsych.2014.08.025.
………………………………………………………
Microarray
https://www.genomicseducation.hee.nhs.uk/genotes/knowledge-hub/microarray-array-cgh/
………………………………………………………
Polygenetic risk scores
Nat Ment Health . 2023 March ; 1(3): 210–223. doi:10.1038/s44220-023-00034-y.
Multivariate genome-wide association meta-analysis of over 1 million subjects identifies loci underlying multiple substance use disorders
TWAS
https://www.nature.com/articles/s42003-023-05279-y
………………………………………………………..
eQTL
Precision Medicine and Pharmacogenetics: What Does Oncology Have that Addiction Medicine Doesn’t?
Henry R. Kranzler, M.D.Rachel V. Smith, M.P.H., Ph.D. Robert Schnoll, Ph.D, Afaf Moustafa1,4, and Emma Greenstreet-Akman
Addiction . 2017 December ; 112(12): 2086–2094. doi:10.1111/add.13818.
Toward Precision Medicine in Addiction Treatment
Nora Volkow, MD
The American Journal on Addictions, 27: 35–36, 2018 Copyright © 2017 American Academy of Addiction Psychiatry
ISSN: 1055-0496 print / 1521-0391 online
DOI: 10.1111/ajad.12666
…………………………………………………………….
Jeff Kay 03/2026
Sober Synthesis 
Leave a comment