2021
DOI: 10.1101/2021.05.28.446066
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

A Biologically Interpretable Graph Convolutional Network to Link Genetic Risk Pathways and Neuroimaging Markers of Disease

Abstract: We propose a novel deep neural network for whole-genome imaging-genetics. Our genetics module uses hierarchical graph convolution and pooling operations that mimic the organization of a well-established gene ontology to embed subject-level data into a latent space. The ontology implicitly tracks the convergence of genetic risk across biological pathways, and an attention mechanism automatically identifies the salient edges in our network. We couple the imaging and genetics data using an autoencoder and predict… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
2
1
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(2 citation statements)
references
References 50 publications
0
2
0
Order By: Relevance
“…There are several studies investigating how to use machine learning approaches to explain biological mechanisms. Biologically informed deep learning has shown promising explainability in previous studies 30,31 . Elmarakeby et al 30 designed biologically informed paths to constrain the connection between neurons in the fullyconnected neural network.…”
Section: Introductionmentioning
confidence: 98%
See 1 more Smart Citation
“…There are several studies investigating how to use machine learning approaches to explain biological mechanisms. Biologically informed deep learning has shown promising explainability in previous studies 30,31 . Elmarakeby et al 30 designed biologically informed paths to constrain the connection between neurons in the fullyconnected neural network.…”
Section: Introductionmentioning
confidence: 98%
“…They derived meaningful paths from the importance score during backpropagation, whereas the fully-connected network could not explain the mechanism behind the interactions between two entities. Ghosal et al 31 presented important pathways of hierarchical biological processes by using an attention-based graph neural network. But they only considered the relation between biological processes, and thus failed to explain the impact of multimodal interactions.…”
Section: Introductionmentioning
confidence: 99%