AttOmics: attention-based architecture for diagnosis and prognosis from omics data

Aurélien Beaude (1) , Milad Rafiee Vahid , Franck Augé , Farida Zehraoui (1) , Blaise Hanczar (1)

1 IBISC - Informatique, BioInformatique, Systèmes Complexes

Motivation The increasing availability of high-throughput omics data allows for considering a new medicine centered on individual patients. Precision medicine relies on exploiting these high-throughput data with machine-learning models, especially the ones based on deep-learning approaches, to improve diagnosis. Due to the high-dimensional small-sample nature of omics data, current deep-learning models end up with many parameters and have to be fitted with a limited training set. Furthermore, interactions between molecular entities inside an omics profile are not patient specific but are the same for all patients. Results In this article, we propose AttOmics, a new deep-learning architecture based on the self-attention mechanism. First, we decompose each omics profile into a set of groups, where each group contains related features. Then, by applying the self-attention mechanism to the set of groups, we can capture the different interactions specific to a patient. The results of different experiments carried out in this article show that our model can accurately predict the phenotype of a patient with fewer parameters than deep neural networks. Visualizing the attention maps can provide new insights into the essential groups for a particular phenotype. Availability and implementation The code and data are available at https://forge.ibisc.univ-evry.fr/abeaude/AttOmics. TCGA data can be downloaded from the Genomic Data Commons Data Portal.

Domains: Artificial Intelligence [cs.AI]

Aurélien Beaude, Milad Rafiee Vahid, Franck Augé, Farida Zehraoui, Blaise Hanczar. AttOmics: attention-based architecture for diagnosis and prognosis from omics data. Intelligent Systems for Molecular Biology, Jun 2023, lyon, France. pp.i94-i102, ⟨10.1093/bioinformatics/btad232⟩. ⟨hal-04431940⟩