Multi-omics integration and batch correction using a modality-agnostic deep learning framework
Authors:
- Jose Ignacio Alvira LarizgoitiaJLJose Ignacio Alvira Larizgoitia
- Gabriele PartelGPGabriele Partel
- Lorenzo VenturelliLVLorenzo Venturelli
- Wanqiu ZhangWZWanqiu Zhang
- Xander SpotbeenXSXander Spotbeen
- Sebastiaan VanuytvenSVSebastiaan Vanuytven
- Sam KintSKSam Kint
- Katy VandereykenKVKaty Vandereyken
- David WoutersDWDavid Wouters
- Anis IsmailAIAnis Ismail
- Regis ScarceriauxRSRegis Scarceriaux
- Jakub IdkowiakJIJakub Idkowiak
Preprint
Abstract
State-of-the-art biotechnologies allow the detection of different molecular species on the same biological sample, generating complex highly-dimensional multi-modal datasets. Gaining a holistic understanding of biological phenomena, such as oncogenesis or aging, requires integrating these diverse modalities into low-dimensional data representations while correcting for technical artifacts. Here we present MIMA, a modular, unsupervised AI framework for multi-omics data integration and batch correction. Applied to complex spatial and single-cell datasets, MIMA effectively removes batch effects, while preserving biologically relevant information, and learns representations predictive of expert pathologist annotations. Additionally, it enables cross-modal translation, uncovers molecular patterns not captured by manual annotations, and despite being modality-agnostic performs on par with specialized state-of-the-art tools. MIMA’s flexibility and scalability make it a powerful tool for multimodal data analysis. MIMA provides a foundation for AI-based, multi-omics augmented digital pathology frameworks, offering new opportunities for improved patient stratification and precision medicine through the comprehensive integration of high-dimensional molecular data and histopathological imaging.Keywords
Funding Information
- FWO (1SF1522N)
- FWO (1SF1524N)
- FWO (1S49218N)
- FWO (1S93320N)
- FWO (S001623N)
- FWO (S005319N)
- Marie Skłodowska-Curie Actions Doctoral Network (101120283)
- VLAIO (HBC.20192204)
- Australian Research Council (FT190100082)
- NCI (DoD HT9425-25-1-0385)
- NCI (P01 CA265768)
- EU HORIZON-MSCA-DN (101120283)
- FWOEOS (30837538)
- FWOEOS (G0F6718N)
This publication has 34 references indexed in Scilit:
- Integrating Ambient Ionization Mass Spectrometry Imaging and Spatial Transcriptomics on the Same Cancer Tissues to Identify RNA–Metabolite CorrelationsAngewandte Chemie International Edition, 2025
- The technological landscape and applications of single-cell multi-omicsNature Reviews Molecular Cell Biology, 2023
- Stabilized mosaic single-cell data integration using unshared featuresNature Biotechnology, 2023
- Methods and applications for single-cell and spatial multi-omicsNature Reviews Genetics, 2023
- scMoMaT jointly performs single cell mosaic integration and multi-modal bio-marker detectionNature Communications, 2023
- Computational Approaches and Challenges in Spatial TranscriptomicsGenomics, Proteomics & Bioinformatics, 2022
- Computational principles and challenges in single-cell data integrationNature Biotechnology, 2021
- Batch Effects in MALDI Mass Spectrometry ImagingJournal of the American Society for Mass Spectrometry, 2021
- Harmonization of quality metrics and power calculation in multi-omic studiesNature Communications, 2020
- Simultaneous epitope and transcriptome measurement in single cellsNature Methods, 2017