Browsing by Subject "Ancestry"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Ancestry May Confound Genetic Machine Learning: Candidate-Gene Prediction of Opioid Use Disorder as an Example
    (Elsevier, 2021) Hatoum, Alexander S.; Wendt, Frank R.; Galimberti, Marco; Polimanti, Renato; Neale, Benjamin; Kranzler, Henry R.; Gelernter, Joel; Edenberg, Howard J.; Agrawal, Arpana; Medical and Molecular Genetics, School of Medicine
    Background: Machine learning (ML) models are beginning to proliferate in psychiatry, however machine learning models in psychiatric genetics have not always accounted for ancestry. Using an empirical example of a proposed genetic test for OUD, and exploring a similar test for tobacco dependence and a simulated binary phenotype, we show that genetic prediction using ML is vulnerable to ancestral confounding. Methods: We utilize five ML algorithms trained with 16 brain reward-derived "candidate" SNPs proposed for commercial use and examine their ability to predict OUD vs. ancestry in an out-of-sample test set (N = 1000, stratified into equal groups of n = 250 cases and controls each of European and African ancestry). We rerun analyses with 8 random sets of allele-frequency matched SNPs. We contrast findings with 11 genome-wide significant variants for tobacco smoking. To document generalizability, we generate and test a random phenotype. Results: None of the 5 ML algorithms predict OUD better than chance when ancestry was balanced but were confounded with ancestry in an out-of-sample test. In addition, the algorithms preferentially predicted admixed subpopulations. Random sets of variants matched to the candidate SNPs by allele frequency produced similar bias. Genome-wide significant tobacco smoking variants were also confounded by ancestry. Finally, random SNPs predicting a random simulated phenotype show that the bias attributable to ancestral confounding could impact any ML-based genetic prediction. Conclusions: Researchers and clinicians are encouraged to be skeptical of claims of high prediction accuracy from ML-derived genetic algorithms for polygenic traits like addiction, particularly when using candidate variants.
  • Thumbnail Image
    Item
    A genetic investigation into a Lebanese population: from STR’s to SNP’s
    (2018-06-26) Ghemrawi, Mirna; Walsh, Susan
    In the past, the present and the future, Lebanon has been an important link between the East and the West. It was always known as the ‘Switzerland of the East’. Over the years, it was a hotspot for different civilizations that uniquely shaped the genomic backbone of the current Lebanese. It is also a good representation of genetically admixed individuals with diverse phenotype characteristics and unique features. Lebanon, quite like other Middle Eastern populations, lacks sufficient genetic studies that helps to better comprehend the complex genomic composition of different traits and diseases. The lack of good representation of the Middle East and North Africa (MENA) region in global studies has led to ambiguity in discovering special ancestry markers and patterns in the Lebanese genome. Yet, in this study, a thorough investigation into a Lebanese collection shows new patterns that potentially would be helpful in forensic and genealogical applications. The investigation into the autosomal and Y-STRs revealed unique alleles that would be valuable in future forensic investigation analysis. In addition, the assessment of phenotype prediction models to predict eye, hair and skin color showed promising results in terms of prediction performance. Those results encourage the future use of intelligence tools in the regions that in return would aid in serving justice and furthering science research. In fact, ancestry and genetic distance studies confirms the presence of admixture within Lebanon between Europe and North Africa.
  • Thumbnail Image
    Item
    Indy Postcard Collector, February 2023
    (Indianapolis Postcard Club, 2023-02) Hook , Sara Anne
    The February 2023 issue of Indy Postcard Collector, published by the Indianapolis Postcard Club, edited by Sara Anne Hook, Professor Emerita.
  • Thumbnail Image
    Item
    Metabolic Links to Socioeconomic Stresses Uniquely Affecting Ancestry in Normal Breast Tissue at Risk for Breast Cancer
    (Frontiers Media, 2022-06-27) Rujchanarong, Denys; Scott, Danielle; Park, Yeonhee; Brown, Sean; Mehta, Anand S.; Drake, Richard; Sandusky, George E.; Nakshatri, Harikrishna; Angel, Peggi M.; Pathology and Laboratory Medicine, School of Medicine
    A primary difference between black women (BW) and white women (WW) diagnosed with breast cancer is aggressiveness of the tumor. Black women have higher mortalities with similar incidence of breast cancer compared to other race/ethnicities, and they are diagnosed at a younger age with more advanced tumors with double the rate of lethal, triple negative breast cancers. One hypothesis is that chronic social and economic stressors result in ancestry-dependent molecular responses that create a tumor permissive tissue microenvironment in normal breast tissue. Altered regulation of N-glycosylation of proteins, a glucose metabolism-linked post-translational modification attached to an asparagine (N) residue, has been associated with two strong independent risk factors for breast cancer: increased breast density and body mass index (BMI). Interestingly, high body mass index (BMI) levels have been reported to associate with increases of cancer-associated N-glycan signatures. In this study, we used matrix assisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS) to investigate molecular pattern changes of N-glycosylation in ancestry defined normal breast tissue from BW and WW with significant 5-year risk of breast cancer by Gail score. N-glycosylation was tested against social stressors including marital status, single, education, economic status (income), personal reproductive history, the risk factors BMI and age. Normal breast tissue microarrays from the Susan G. Komen tissue bank (BW=43; WW= 43) were used to evaluate glycosylation against socioeconomic stress and risk factors. One specific N-glycan (2158 m/z) appeared dependent on ancestry with high sensitivity and specificity (AUC 0.77, Brown/Wilson p-value<0.0001). Application of a linear regression model with ancestry as group variable and socioeconomic covariates as predictors identified a specific N-glycan signature associated with different socioeconomic stresses. For WW, household income was strongly associated to certain N-glycans, while for BW, marital status (married and single) was strongly associated with the same N-glycan signature. Current work focuses on understanding if combined N-glycan biosignatures can further help understand normal breast tissue at risk. This study lays the foundation for understanding the complexities linking socioeconomic stresses and molecular factors to their role in ancestry dependent breast cancer risk.
  • Thumbnail Image
    Item
    The genetic architecture of pediatric cardiomyopathy
    (Elsevier, 2022) Ware, Stephanie M.; Bhatnagar, Surbhi; Dexheimer, Phillip J.; Wilkinson, James D.; Sridhar, Arthi; Fan, Xiao; Shen, Yufeng; Tariq, Muhammad; Schubert, Jeffrey A.; Colan, Steven D.; Shi, Ling; Canter, Charles E.; Hsu, Daphne T.; Bansal, Neha; Webber, Steven A.; Everitt, Melanie D.; Kantor, Paul F.; Rossano, Joseph W.; Pahl, Elfriede; Rusconi, Paolo; Lee, Teresa M.; Towbin, Jeffrey A.; Lal, Ashwin K.; Chung, Wendy K.; Miller, Erin M.; Aronow, Bruce; Martin, Lisa J.; Lipshultz, Steven E.; Pediatric Cardiomyopathy Registry Study Group; Pediatrics, School of Medicine
    To understand the genetic contribution to primary pediatric cardiomyopathy, we performed exome sequencing in a large cohort of 528 children with cardiomyopathy. Using clinical interpretation guidelines and targeting genes implicated in cardiomyopathy, we identified a genetic cause in 32% of affected individuals. Cardiomyopathy sub-phenotypes differed by ancestry, age at diagnosis, and family history. Infants < 1 year were less likely to have a molecular diagnosis (p < 0.001). Using a discovery set of 1,703 candidate genes and informatic tools, we identified rare and damaging variants in 56% of affected individuals. We see an excess burden of damaging variants in affected individuals as compared to two independent control sets, 1000 Genomes Project (p < 0.001) and SPARK parental controls (p < 1 × 10-16). Cardiomyopathy variant burden remained enriched when stratified by ancestry, variant type, and sub-phenotype, emphasizing the importance of understanding the contribution of these factors to genetic architecture. Enrichment in this discovery candidate gene set suggests multigenic mechanisms underlie sub-phenotype-specific causes and presentations of cardiomyopathy. These results identify important information about the genetic architecture of pediatric cardiomyopathy and support recommendations for clinical genetic testing in children while illustrating differences in genetic architecture by age, ancestry, and sub-phenotype and providing rationale for larger studies to investigate multigenic contributions.