Zartaloudi, E; (2021) Identification and Evaluation of Endophenotypes and Biomarkers of Schizophrenia and Bipolar Disorder: Genomic Dissection of the Psychosis Phenotype. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Zartaloudi_10125568_thesis_sig_removed.pdf Download (2MB) | Preview

Abstract

Background: Psychotic disorders affect approximately 3% of the population. Over 100 genetic variants have been associated with schizophrenia and about 50 with bipolar disorder. Each of them individually has a small effect on disease risk but combined in a cumulative polygenic risk score (PRS), they have a major impact. Copy number variants (CNVs) have also been associated with schizophrenia. However, little is known about their functional effects. The investigation of endophenotypes, which fall in the genotype to phenotype pathway, could help us understand the role of genetic variants and their mechanisms. Methods: In chapter 1 of my thesis, I reviewed the literature on endophenotypes, and genetic variants associated with psychosis, which revealed that the interrelationships between several well-established cognitive, neuroimaging and electrophysiological psychosis endophenotypes, and the joint contributions of CNV burden and polygenic risk scores on psychosis risk have not been studied yet. I investigated those topics in chapters 3 and 4 respectively. In chapter 2 I carried out a scoping review of CNVs associated with neurodevelopmental disorders, psychosis and cognition and carried out a meta-analysis of 16p11.2 distal deletion in schizophrenia. I also investigated the influences of CNV size on schizophrenia risk for 53 CNVs. For all the analyses, I used CNVcatalog, which is a new repository me and my supervisors created, incorporating data from published studies examining associations of CNV loci with several clinical phenotypes, including schizophrenia. Finally, in chapter 5 I summarise the main findings of my thesis and I discuss the strengths, limitations and clinical implications of my research. Results: Chapter 2: The meta-analysis of 16p11.2 distal deletion in schizophrenia revealed that carriers of that CNV had higher risk of developing schizophrenia compared to non carriers. I also found that larger CNV size was associated with larger effect sizes when examining all CNVs together (both deletions and duplications) and CNV deletions. However, the size was not significanly associated with disease risk for CNV duplications. Chapter 3: All the cognitive endophenotypes were associated with each other. Endophenotypes across imaging, cognitive and electrophysiological domains did not show a correlation. The relationships between pairs of endophenotypes were consistent in all three participant groups (cases with psychosis, their unaffected relatives and healthy controls), differing for some of the cognitive pairings only in the strengths of the relationships. Chapter 4: I examined the joint contributions of CNV burden and polygenic risk scores on psychosis risk. I analysed two datasets separately and then combined them by meta-analysis. CNV burden and PRS could explain 11.8% and 10.8% of the variance in disease risk in each dataset. The classification accuracy of my models was 81%, 83% and 77% for the comparisons of all psychosis cases vs controls, schizophrenia cases vs controls and bipolar cases vs controls respectively. The addition of CNV burden to the models increased the variance explained only by 0.1% for MPL dataset and by 0.08% in the PEIC dataset. Discussion: Findings from my thesis contribute to our current knowledge on psychosis endophenotypes and on the genetic influences in psychoses. Deciphering the genetic architecture of psychotic disorders could hopefully in the future improve the lives of affected individuals.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item