Revealing the relationship between human genome regions and pathological phenotypes through network analysis

Abstract

Recent advances in sequencing technologies allow researchers to investigate diseases resulting of genomic variation. This allows us to further develop the concept of precision medicine and determine the best treatment for each patient. We have focused on developing tools for studying genomic loci associated to pathological traits from the perspective of network analysis. We have obtained from DECIPHER database patient information which includes their affected genomic regions by Copy Number Variations (CNV) and their pathologies described as Human Phenotype Ontology phenotypes. We have used different metrics for calculating association values between phenotypes and affected genomic regions to determine which method fits better to our data. The results obtained in this work, can be used in prediction systems for determining and ranking which genomic regions are associated to a concrete phenotype, in order to help clinicians with their diagnosis.