Abstract
Untangling the complex interplay between phenotype and genotype is crucial to the effective characterization and subtyping of diseases. Here we build and analyze the multiplex network of 779 human diseases, which consists of a genotype-based layer and a phenotype-based layer. We show that diseases with common genetic constituents tend to share symptoms, and uncover how phenotype information helps boost genotype information. Moreover, we offer a flexible classification of diseases that considers their molecular underpinnings alongside their clinical manifestations. We detect cohesive groups of diseases that have high intra-group similarity at both the molecular and the phenotypic level. Inspecting these disease communities, we demonstrate the underlying pathways that connect diseases mechanistically. We observe monogenic disorders grouped together with complex diseases for which they increase the risk factor. We propose potentially new disease associations that arise as a unique feature of the information flow within and across the two layers.
Highlights
The advent of next-generation sequencing (NGS) and genomewide association studies (GWAS) has led to the accumulation of a vast amount of disease-gene associations.[1]
Our study represents a novel addition to the body of works addressing this subject in two ways: (1) We provide a large-scale multiplex disease network, which has not previously been constructed consistently using multiple aligned data sources; and (2) We apply, for the first time, a multiplex community detection method on a global disease network for the purposes of disease classification, in contrast to other similar methods that have been used on multilayered molecular networks to determine functional similarities between biological molecules
We identified the communities by using Infomap, a well-known algorithm based on the compression of information flow.[38,39]
Summary
The advent of next-generation sequencing (NGS) and genomewide association studies (GWAS) has led to the accumulation of a vast amount of disease-gene associations.[1]. Connecting disease-related phenotypes to their underlying molecular mechanisms and genetic constituents is crucial for a better understanding of complex human diseases. The emerging field of network medicine offers the tools of network science for distilling relevant insight from the growing sets of molecular disease omics data.[2] One of the earliest attempts at exploring the higher-level implications of disease-gene associations from the network perspective was the construction of genotype-based disease networks, useful to show the global organization of diseases around functional modules[3] and to infer comorbidity relations between diseases.[4] On the same basis, phenotypic–based disease networks were constructed by text-mining large-scale Medicare data, systematically classifying diseases based on phenotype similarity,[5] and facilitating the identification of patterns of disease progression.[6] Since these pioneering works, many studies have focused on adding to the growing compendium of diseasedisease associations. Menche et al identified common mechanistic pathways between diseases by the overlap of disease modules.[9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
