An Investigation Into the Utility of Guilt by Association Machine Learning Algorithms for the Prioritization of Autism Spectrum Disorder Candidate Risk Genes

An Investigation Into the Utility of Guilt by Association Machine Learning Algorithms for the Prioritization of Autism Spectrum Disorder Candidate Risk Genes

An investigation into the utility of guilt by association machine learning algorithms for the prioritization of autism spectrum disorder candidate risk genes by Margot Patricia Rainbow Gunning B.Sc.H, Queen’s University, 2017 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Bioinformatics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) December 2019 © Margot Patricia Rainbow Gunning, 2019 The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled: An investigation into the utility of guilt by association machine learning algorithms for the prioritization of autism spectrum disorder risk genes submitted Margot Patricia Rainbow in partial fulfillment of the requirements by Gunning for the degree of Master of Science in Bioinformatics Examining Committee: Paul Pavlidis, Michael Smith Laboratories/Psychiatry Supervisor Kurt Haas, Cellular and Physiological Sciences Supervisory Committee Member Elodie Portales-Casamar, Clinical Research Informatics Lead, BCCHR Supervisory Committee Member Joerg Gsponer, Biochemistry/Molecular Biology Additional Examiner ii Abstract Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by impairments in social interaction and communication, and restrictive repetitive behaviours or interests, with extreme phenotypic and genetic heterogeneity. Currently, genetic association studies have identified 90 risk genes with high confidence out of an estimated 1000. Researchers have begun to use machine learning methods leveraging heterogeneous biological network data in attempts to aid in discovery of ASD risk genes. However, the real-world utility of these studies is questionable: network-based machine learners are often biased towards well studied genes because they operate on a principle called “guilty by association.” In this thesis, I evaluate and compare genetic and computation approaches to ASD risk gene prioritization. I demonstrate that network-based computational approaches are adding little additional useful information compared to genetic approaches for prioritization. Furthermore, I demonstrate that gene expression profiles, and generic measures of disease gene likelihood may provide less biased contextual information that can be used to supplement genetic association data to prioritize ASD risk genes. Lastly, I discuss how data quality and data dependence impacts evaluation of machine learning algorithms and genetic association studies. iii Lay Summary Autism spectrum disorder (ASD) is an extremely heterogeneous neurodevelopmental disorder associated with social and communication deficits. Over the past decades, inroads have been made into delineating the complex genetic nature of ASD. However, we still have much to learn as only a fraction of the estimated genetic risk factors have been identified with high confidence. In this thesis, I evaluate different types of ASD gene discovery and prioritization methods, including genetic and computation approaches, to answer the question: can use of other non-genetic types of biological data aid in ASD gene discovery? iv Preface I was responsible for all word described, under the direction of my thesis supervisor, Dr. Paul Pavlidis, with the following exceptions: Manuel Belmadani was responsible for setting up the Ensembl Variant Effect Predictor used in Chapter 2. Shams Bhuiyan and Nathaniel Lim provided me with gene level annotation files for publication numbers and multifunctionality rankings used in Chapters 2 and 3. v Table of Contents Abstract ......................................................................................................................................... iii Lay Summary ............................................................................................................................... iv Preface ............................................................................................................................................ v Table of Contents ......................................................................................................................... vi List of Tables ............................................................................................................................... vii List of Figures ............................................................................................................................. viii List of Abbreviations ................................................................................................................... ix Acknowledgements ....................................................................................................................... x Dedication .................................................................................................................................... xi Chpter 1: Introduction ................................................................................................................. 1 1.1: Motivation .................................................................................................................... 1 1.2: What is ASD? .............................................................................................................. 3 1.3: Characterizing and predicting the impact of human genetic variation ........................ 6 1.4: The heterogeneous genetic etiology of ASD ............................................................. 16 1.5: Machine learning for prediction of autism spectrum disorder genes......................... 23 1.6: Thesis outline ............................................................................................................. 32 Chapter 2: Collecting information on current ASD candidate genes .................................... 33 2.1: Introduction ................................................................................................................ 33 2.2: Materials and Methods............................................................................................... 34 2.3: Results ........................................................................................................................ 50 2.4: Discussion .................................................................................................................. 57 Chapter 3: Evaluation of ASD gene prioritization studies...................................................... 61 3.1: Introduction ................................................................................................................ 61 3.2: Materials and Methods............................................................................................... 62 3.3: Results ........................................................................................................................ 82 3.4: Discussion ................................................................................................................ 110 Chapter 4: Conclusions ............................................................................................................ 121 Bibliography .............................................................................................................................. 124 Appendix A: Additional information for Chapter 2................................................................... 145 Appendix B: Additional information for Chapter 3 ................................................................... 153 vi List of Tables Table 2.1: ASD risk gene sets ...................................................................................................... 34 Table 2.2: Examples of variants reported in SFARI with non or inconsistent HGVS format ..... 37 Table 2.3: Example of incomplete information in ClinVar ......................................................... 38 Table 2.4: Examples of unresolved inheritance pattern reported in the same sample across multiple studies ............................................................................................................................. 40 Table 2.5: Example of the different VEP annotations resulting from different variant input formats. ......................................................................................................................................... 41 Table 2.6: Data included in the GRCh37 VEP cache .................................................................. 43 Table 2.7: Plugins installed for VEP and utilized for annotation ................................................ 44 Table 2.8: VEP hierarchy of variant consequences ..................................................................... 45 Table 2.9: Gene and Variant collection statistics ......................................................................... 51 Table 2.10: Median values of constraint scores for ASD gene sets ............................................. 54 Table 2.11: Median values of number of publications, multifunctionality and physical node degree for gene sets. ...................................................................................................................... 56 Table 3.1: Gene sets used for evaluation ..................................................................................... 76 Table 3.2: Genetic association and GBA-based ML ASD gene prioritization studies. ............... 76 Table 3.3: AUROC, Precision

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    167 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us