Jesper N. Tegnér

Unify learns cellular evolution with universal multimodal embeddings

Integrating single-cell RNA-sequencing (scRNA-seq) data across species is hindered by evolutionary divergence, technical batch effects, and the reliance on one-to-one orthologs. Here, we present Unify, a transfer learning methodology that learns universal cell embeddings by defining functionally coherent, multi-modal macrogenes. This is achieved by combining RNA expression with embeddings from protein language models and general-purpose language models. Unify transcends species boundaries, enabling cross-species comparisons beyond strict gene-level homology. Unify corrects batch effects while preserving conserved biological signals across vast evolutionary distances and enables more accurate prediction of perturbation responses across species, such as from mouse to human. Applied to species separated by over 700 million years, Unify reconstructs more accurate multi-species cell-type evolutionary trees and uncovers convergent gene programs. Together, these results establish Unify as a powerful method for comparative single-cell genomics and evolutionary biology. Integrating single-cell RNA-sequencing (scRNA-seq) data across species is still technically challenging. Here, the authors report a transfer learning framework designed to integrate scRNA-seq data across species by combining RNA expression with embeddings from protein language models and general-purpose language models.

Single-cell analysis suggests coordinated immune-cell redistribution between blood and cerebrospinal fluid in early multiple sclerosis

Despite immune-cell infiltration being a hallmark of multiple sclerosis (MS), the interplay between the periphery and central nervous system immune responses is still incompletely characterized. We performed single-cell transcriptomic and V(D)J sequencing of paired blood and cerebrospinal fluid (CSF) immune cells from treatment-naive relapsing-remitting women with MS and compared them to age– and sex-matched healthy controls. Across major immune lineages, we identified coordinated, compartment-specific immune alterations, with enrichment of activated and memory lymphocyte populations in the CSF and concomitant depletion of related populations in peripheral blood, suggesting their recruitment from blood to CSF. Clonally expanded CD4 memory T-cells, together with activated, expanded IgM-positive B-cells, accumulated predominantly in the CSF of people with MS compared to healthy subjects. Interestingly, tissue-primed cytotoxic populations and CXCR3-associated memory populations were depleted from the CSF of MS, implying their recruitment to the target tissue in early disease. These findings reveal coordinated, compartment-specific immune changes in early MS and provide a systems-level view of immune-cell trafficking between peripheral and central nervous system compartments.

The human RNA-DNA interactome is cell type-specific and dynamic