HoloCell: New Foundation Model Integrates Three Layers of Cellular Data
Researchers have developed HoloCell, a generative AI foundation model with over 860 million parameters designed to jointly model all three major single-cell omics modalities — epigenomics, transcriptomics, and proteomics. The model was pretrained on approximately 468 million single-cell profiles comprising over 425 billion tokens, using a hierarchical tokenization strategy grounded in biological structure. HoloCell advances the concept of a 'virtual cell' by enabling both unified representation of cellular states and generative simulation of multi-omics information flow.
HoloCell is presented as the first generative foundation model capable of joint representation learning and generative modeling across epigenomics, transcriptomics, and proteomics simultaneously. Trained on the Human-Multi-Omics-Corpus — a dataset of roughly 468 million single-cell profiles — the model encodes cis-regulatory elements, genes, and proteins as structured tokens within a shared framework using a biologically grounded hierarchical tokenization approach. A key technical challenge it addresses is robustness to heterogeneous modality missingness, meaning it can function even when not all omics layers are available for a given cell. The model was evaluated across single-omics representation learning, paired and unpaired multi-omics integration, and cross-modal generation using an iterative diffusion and remasking mechanism that accounts for the unordered nature of biological features. The authors report superior performance and flexibility across these diverse tasks compared to existing approaches. By combining systematic cellular characterization with generative simulation capabilities, HoloCell is positioned as a step toward a computational 'virtual cell' — a long-standing goal in systems biology.
What's missing
As a preprint on bioRxiv, HoloCell has not yet undergone formal peer review. The claim of being 'the first' such model has not been independently verified.
What different sources said
- bioRxivCenter
HoloCell: A Generative Foundation Model for Holistic Cellular Modeling
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.