Unveiling new disease, pathway, and gene associations via multi-scale neural network

Abstract

Diseases involve complex modifications to the cellular machinery. The gene expression profile of the affected cells contains characteristic patterns linked to a disease. Hence, new biological knowledge about a disease can be extracted from these profiles, improving our ability to diagnose and assess disease risks. This knowledge can be used for drug re-purposing, or by physicians to evaluate a patient's condition and co-morbidity risk.Here, we consider differential gene expressions obtained by microarray technology for patients diagnosed with various diseases. Based on these data and cellular multi-scale organization, we aim at uncovering disease-disease links, disease-gene and disease-pathways associations. We propose neural networks with structures based on the multi-scale organization of proteins in a cell into protein complexes and pathways. We show that these models are able to correctly predict the diagnosis for the majority of patients. Through the analysis of the trained models, we predict disease-disease, disease-pathway, and disease-gene associations and validate the predictions by comparisons to known interactions and literature search, proposing putative explanations for the predictions.

Deriving disease modules from the compressed transcriptional space embedded in a deep autoencoder

“…Also, the presented approach can play a crucial role in utilizing the resolution level of the single cell transcriptomic signals in prioritizing genes that are enriched with the upstream dysregulated genes and their relationship with causal genetic variants 37 . Another important application of our approach can indeed provide new insights in the multiscale organization about disease-disease, disease pathways disease-gene associations 38 .…”

Deriving Disease Modules from the Compressed Transcriptional Space Embedded in a Deep Auto-encoder

“…Another important application of our approach can indeed provide new insights in the multiscale organization about disease-disease, disease pathways disease-gene associations (Gaudelet et al, 2019).…”

Building an ecologically founded disease risk prioritization framework for migratory wildlife species based on contact with livestock

“…Even though existing frameworks build up from individuals to populations (eg. Garabed et al, 2019;Garira et al, 2020), often they consider only one definitive host (Morgan et al, 2004), employ resource-intense methodologies (Gaudelet et al, 2020), and fail to incorporate both the social and ecological aspects driving potential disease spill-over risk across various ecological scales. Additionally, Schwartz et al (2018) caution against using any one framework in isolation as it risks diminishing potential benefits, as no one framework covers the full spectrum of potential conservation planning and decision challenges.…”