Mechanism-aware and multimodal AI: beyond model-agnostic interpretation

Annalisa Occhipinti; Suraj Verma; Le Minh Thao Doan; Claudio Angione

doi:10.1016/j.tcb.2023.11.002

Mechanism-aware and multimodal AI: beyond model-agnostic interpretation

Trends Cell Biol. 2024 Feb;34(2):85-89. doi: 10.1016/j.tcb.2023.11.002. Epub 2023 Dec 11.

Authors

Annalisa Occhipinti¹, Suraj Verma², Le Minh Thao Doan², Claudio Angione³

Affiliations

¹ School of Computing, Engineering and Digital Technologies, Teesside University, Middlesborough, UK; Centre for Digital Innovation, Teesside University, Middlesborough, UK; National Horizons Centre, Teesside University, Darlington, UK.
² School of Computing, Engineering and Digital Technologies, Teesside University, Middlesborough, UK.
³ School of Computing, Engineering and Digital Technologies, Teesside University, Middlesborough, UK; Centre for Digital Innovation, Teesside University, Middlesborough, UK; National Horizons Centre, Teesside University, Darlington, UK. Electronic address: c.angione@tees.ac.uk.

PMID: 38087709
DOI: 10.1016/j.tcb.2023.11.002

Abstract

Artificial intelligence (AI) is widely used for exploiting multimodal biomedical data, with increasingly accurate predictions and model-agnostic interpretations, which are however also agnostic to biological mechanisms. Combining metabolic modelling, 'omics, and imaging data via multimodal AI can generate predictions that can be interpreted mechanistically and transparently, therefore with significantly higher therapeutic potential.

Keywords: imaging data; mechanism-aware AI; metabolic modelling; multi-omics; multimodal integration.

MeSH terms

Artificial Intelligence*
Models, Biological
Multiomics