Structured quantification of inherited macular disease phenotypes as the basis for automated algorithms to determine causal genes

Columbia University Health Sciences, New York NY

Investigators

Abstract

PROJECT SUMMARY/ABSTRACT The ability to obtain a molecular diagnosis in patients by genetic testing is an essential step towards understanding the etiology and treatment prospects for a monogenic disease. Recently, genetic testing has become increasingly accessible for patients due to technical advances in sequencing, however, genetic testing outcomes in many cases are often inconclusive. One major underlying reason for this is that identifying the precise causal gene of a monogenic disorder based solely on the interpretation of variant pathogenicity is challenging due to the profound complexity of the human genome and its mutational landscape. The corresponding phenotypic landscape in patients exhibits just as much (if not more) complexity however these clinical insights are underutilized in the pursuit of molecular diagnoses. The overall goal of this proposal is to characterize critical differences in the clinical manifestations of various inherited macular dystrophy (IMD) genes and develop a molecular diagnosis prediction tool to improve genetic testing outcomes. In Specific Aim 1, we will assemble a large multi-IMD gene cohort of patients in which we will acquire a comprehensive set of phenotypic data using advanced retinal imaging modalities. The causal gene in each patient will be confirmed by whole exome sequencing (WES) and analysis using a modified diagnostic pipeline specifically tailored to IMD genes. In Specific Aim 2, we will (1) evaluate differences in the pattern of retinal pigment epithelium loss relative to photoreceptor loss in the degenerating macular lesions of each eye using comparative, cell type-specific fundus autofluorescence imaging, (2) measure early sub-clinical photoreceptor layer thinning across the macula with transverse swept-source optical coherence tomography and (3) compare morphometric characteristics of sub- retinal fundus deposition found across various IMD genes. In Specific Aim 3, we will identify and determine ways to statistically differentiate “diagnostically ambiguous phenotypes” (DAP) shared between IMD genes and integrate these findings into a computational model that predicts and algorithmically ranks candidate genes based on fundus findings in an individual patient. Lastly, this model will be tested and validated in a separate IMD cohort to evaluate its efficacy in aiding genetic testing in a standard clinical setting.