Network-based Nonparametric Tests to Identify Genetic Modifiers of Rare Diseases
presentationposted on 10.03.2020 by Eliatan Niktab, Stephen Sturley, Ingrid Winship, Mark Walterfang, Paul Atkinson, Andrew Munkacsi
Academic presentations can be uploaded in their original slide format. Presentations are usually represented as slide decks. Videos of presentations can be uploaded as media.
Genome and exome sequencing has been extensively successful in identifying disease causing gene mutations and variants in GWAS. However, there has been little success in deducing the pertinent genomic variants that significantly modify disease progression and fully account for phenotype. One reason is that current use of genome-wide association study (GWAS) utilize narrow sense heritability estimates and do not include assessment of epistasis and other components of broad-sense heritability1-2 . Here we report investigation of genetic variants that modify the causal gene of a monogenic disease and ultimately regulate its onset and progression in individuals. Niemann-Pick type C (NP-C) disease, a rare monogenic Mendelian disease, is one of more than 6,000 Mendelian diseases for which there is no cure. Most NP-C patients with the NPC1 gene mutation are diagnosed as late infants and die before or during adolescence, yet survival of some to adulthood provides a testbed for elucidating genes that alleviate the primary mutation. Therefore, we collected whole-genome sequences of pediatric and adult-onset patients. We then developed a pipeline that integrates mathematical models of genetic polymorphisms, augmented Bayesian biological networks, clinical records, and semantic ontologies of GWAS data. The tool that we developed analyzes sequencing data, identifies genome-wide interactions, and has scripts that control for confounding factors using heterogeneous data harmonization and modularity-based clustering. Our approach mitigates the statistical challenge of sample sizes inherent to current GWAS methodology. There are a large number of modifying genes that appear to function in epistatic networks of disease-modifying variants whose genetic effects together explain the heritability NP-C in its various manifestations.
1- Kim, S. et al. Genes with high network connectivity are enriched for disease heritability. Am. J. Hum. Genet. (2019).
2- Escala-Garcia, M. et al. A network analysis to identify mediators of germline-driven differences in breast cancer prognosis. Nat. Commun. (2020).
Keywords: system biology, network, genome, GWAS, Bayesian
Acknowledgments: Ara Parseghian Medical Research Fund
ABOUT THE AUTHOR(S)
Eliatan Niktab - I’m a Ph.D. candidate at Victoria University and a member of Quantitative Methods, Machine Learning, and Functional Genomics group at Genomics England Clinical Interpretation Partnership (GeCIP). I’m trained to investigate diverse, complex and multifeature data including human genomics, proteomics, and metabolomics by developing mathematical models and statistical analyses. My Ph.D. dissertation utilizes such models for investigating the rare neurodegenerative Niemann-Pick type C (NPC) disease. I’m mostly practiced in genome-scale algorithm design, using deep neural networks for genetic variant discovery, Baysian modeling, and GPU-accelerated software development.
Stephen Sturley, Ph.D. Professor Sturley’s group uses a multidisciplinary approach that integrates genetics, biochemistry and cell biology. He is specialized in applying yeast as a model organism to understand human lipid metabolism. Particular emphasis and success of his lab has been attained with regard to cholesterol, sphingolipid and fatty acid homeostasis underlying lipotoxicity with particular reference to neurodegeneration, obesity, and diabetes. Their use of yeast to identify genetic modifiers of recessive disorders such as Niemann- Pick Type C (NPC) resulted in the identification of histone deacetylase inhibitors as a candidate therapy, for which this drug was further tested in murine models of NPC disease and now in clinical trial in human patients.
Mark Walterfang, M.D., Ph.D. FRANZCP Professor Walterfang has significant experience in clinical neuropsychiatry and general adult psychiatry with expertise in managing comorbid psychiatric and neurological disorders, neurodegenerative disorders, neurometabolic disorders, and atypical dementias. Clinical experience is the basis of his success in research, starting with 13 published papers from his Ph.D. that have been cited more than 500 times. He has since published over 130 papers in major psychiatric, metabolic, neurological and neuroimaging journals.
Ingrid Winship, MBChB, MD, Ph.D., FRACP, FACD The focus of her research is the relationship between genotype and phenotype with particular emphasis on human diseases. In the last two decades, her research has helped to frame the questions, define the phenotypes, and via statistical analyses associated genotype and phenotype. At the other end of the translational pipeline, her research has translated the discoveries and knowledge into clinical protocols and policies, which has changed the way patients are treated in medical practice via new drug targets and biomarkers to monitor the onset and progression of the disease.
Paul Atkinson, Ph.D. Professor Atkinson is a cell biologist who has long investigated ER-related events in specification of membrane protein synthesis and transport. His studies have included ER, Golgi and plasma membrane specific glycosylation structure determined by multidimensional NMR. Specific membrane glycoproteins studies utilised model systems including membrane maturing viruses. More recently he has utilized yeast gene knockout libraries to investigate epistatic network contribution to phenotype in ER related events.
Andrew Munkacsi, Ph.D. Munkacsi lab studies the genetics, cell biology, and biochemistry of intracellular lipid transport to identify novel targets to treat the defective transport of cholesterol and sphingolipids underlying human disease. His group uses a unique combination of unbiased, high- throughput systems biology approaches in yeast genomic screens based on synthetic lethality, gene expression, protein localization, and protein-protein interactions, as well as exome and genome sequencing of human patients. Dr. Munkacsi successfully used these genome-wide yeast screens to identify unsuspected and precise targets to treat neurodegenerative diseases such as Alzheimer’s disease and Niemann- Pick Type C (NPC) disease, of which a subset have progressed to clinical trials.