Liam tackles complex multimodal single-cell data integration challenges

Abstract

Multi-omics characterization of single cells holds outstanding potential for profiling gene regulatory states of thousands of cells and their dynamics and relations. How to integrate multimodal data is an open problem, especially when aiming to combine data from multiple sources or conditions containing biological and technical variation. We introduce liam, a flexible model for the simultaneous horizontal and vertical integration of paired single-cell multimodal data. Liam learns a joint low-dimensional representation of two concurrently measured modalities, which proves beneficial when the information content or quality of the modalities differ. Its integration accounts for complex batch effects using a tuneable combination of conditional and adversarial training and can be optimized using replicate information while retaining selected biological variation. We demonstrate liams superior performance on multiple multimodal data sets, including Multiome and CITE-seq data. Detailed benchmarking experiments illustrate the complexities and challenges remaining for integration and the meaningful assessment of its success.