Abstract:The increasing availability of large-scale single-cell datasets has enabled the detailed description of cell states across multiple biological conditions and perturbations. In parallel, recent advances in unsupervised machine learning, particularly in transfer learning, have enabled fast and scalable mapping of these new single-cell datasets onto reference atlases. The resulting large-scale machine learning models however often have millions of parameters, rendering interpretation of the newly mapped datasets challenging. Here, we propose expiMap, a deep learning model that enables interpretable reference mapping using biologically understandable entities, such as curated sets of genes and gene programs. The key concept is the substitution of the uninterpretable nodes in an autoencoder’s bottleneck by labeled nodes mapping to interpretable lists of genes, such as gene ontologies, biological pathways, or curated gene sets, for which activities are learned as constraints during reconstruction. This is enabled by the incorporation of predefined gene programs into the reference model, and at the same time allowing the model to learn de novo new programs and refine existing programs during reference mapping. We show that the model retains similar integration performance as existing methods while providing a biologically interpretable framework for understanding cellular behavior. We demonstrate the capabilities of expiMap by applying it to 15 datasets encompassing five different tissues and species. The interpretable nature of the mapping revealed unreported associations between interferon signaling via the RIG-I/MDA5 and GPCRs pathways, with differential behavior in CD8+ T cells and CD14+ monocytes in severe COVID-19, as well as the role of annexins in the cellular communications between lymphoid and myeloid compartments for explaining patient response to the applied drugs. Finally, expiMap enabled the direct comparison of a diverse set of pancreatic beta cells from multiple studies where we observed a strong, previously unreported correlation between the unfolded protein response and asparagine N-linked glycosylation. Altogether, expiMap enables the interpretable mapping of single cell transcriptome data sets across cohorts, disease states and other perturbations. ### Competing Interest Statement Fabian J. Theis consults for Immunai Inc., Singularity Bio B.V., CytoReason Ltd, and Omniscope Ltd, and has ownership interest in Dermagnostix GmbH and Cellarity.

Exploring Structured Molecular Landscape from Single-Cell Multi-Omics Data by an Explainable Multimodal Model

Development of Multiomics in Situ Pairwise Sequencing (Mip-Seq) for Single-cell Resolution Multidimensional Spatial Omics

Multimodal weakly supervised learning to identify disease-specific changes in single-cell atlases

xSiGra: explainable model for single-cell spatial data elucidation

Biologically Informed Deep Learning to Infer Gene Program Activity in Single Cells

Inferring single-cell spatial gene expression with tissue morphology via explainable deep learning

MorphLink: Bridging Cell Morphological Behaviors and Molecular Dynamics in Multi-modal Spatial Omics

Comprehensive single-cell RNA-seq analysis using deep interpretable generative modeling guided by biological hierarchy knowledge

Explainable multi-view framework for dissecting intercellular signaling from highly multiplexed spatial data

A multi-view latent variable model reveals cellular heterogeneity in complex tissues for paired multimodal single-cell data

Integrated analysis of multimodal single-cell data with structural similarity

MIDAS: a deep generative model for mosaic integration and knowledge transfer of single-cell multimodal data

Explainable AI Methods for Multi-Omics Analysis: A Survey

Vec2image: an explainable artificial intelligence model for the feature representation and classification of high-dimensional biological data by vector-to-image conversion

SPICEMIX: Integrative single-cell spatial modeling of cell identity

DIME: Fine-grained Interpretations of Multimodal Models via Disentangled Local Explanations

Contrastively generative self-expression model for single-cell and spatial multimodal data

MultiSC: a deep learning pipeline for analyzing multiomics single-cell data

Designing and interpreting ‘multi-omic’ experiments that may change our understanding of biology

Decoding cell identity with multi-scale explainable deep learning