Study of the components that determine the applicability of pathogenicity predictors in the clinical setting
[eng] The translation of Next Generation Sequencing (NGS) technologies from the research field to the clinical setting and, specifically, the results obtained in terms of diagnostic yield remain far from expected. This situation is due to our present inability to solve the “variant interpretation pr...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Universidad de Barcelona |
| Repositorio: | Dipòsit Digital de la UB |
| OAI Identifier: | oai:diposit.ub.edu:2445/180944 |
| Acceso en línea: | https://hdl.handle.net/2445/180944 http://hdl.handle.net/10803/672715 |
| Access Level: | acceso abierto |
| Palabra clave: | Bioinformàtica Genòmica Genètica humana Laboratoris clínics Bioinformatics Genomics Human genetics Clinical laboratories |
| Sumario: | [eng] The translation of Next Generation Sequencing (NGS) technologies from the research field to the clinical setting and, specifically, the results obtained in terms of diagnostic yield remain far from expected. This situation is due to our present inability to solve the “variant interpretation problem”, which consists in establishing whether a sequence variant is either pathogenic or neural. In this thesis we have focused on how this problem is addressed by pathogenicity predictors, studying the components that determine the applicability of these tools in the clinical setting. First, we have developed a novel approach to assess pathogenicity predictors in terms of both their performance and their suitability for clinical applications. We present a cost framework for assessing and comparing in silico tools, inspired on the use of cost models applied in different fields, from clinical tests to credit assessment in finance. A virtue of this cost framework is that it takes into account the consequences of downstream medical decisions in a simple fashion. Second, we have studied one of the most important factors limiting the performance of pathogenicity predictors: genetic background. In this part, we have studied the relationship between molecular impact and disease severity in hemophilias A and B, for a specific type of sequence variants: compensated pathogenic deviations (CPDs). We have established, studying a dataset of variants in coagulation factors FVIII and FIX, that the disruptive impact of a mutation is not enough to explain the associated phenotype. In parallel, we have characterized the genetic background of these proteins,describing at the molecular level its potential to generate phenotypic variability. Finally, we have characterized the contribution of in silico pathogenicity predictors to the variants identified in gene sequencing panels, using as a model a panel designed for Primary Immunodeficiency Disease (PID), developed in the Immunology and Autoinflammatory diseases’ groups, at the Vall d’Hebron University Hospital. The results obtained illustrate the limits of in silico tools and also a new way to take genetic background into consideration. |
|---|