ArticleNature methods2026
Evaluating the role of pretraining dataset size and diversity on single-cell foundation model performance.
Article in Nature methods, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 9 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
9 citing papers in PubMed.
- Integration of artificial intelligence and multi-omics for precision medicine.Functional & integrative genomics · 2026Review
- Organism-scale annotation with Pan-human Azimuth.bioRxiv : the preprint server for biology · 2026Article
- Evaluating the learnability of single-cell large language models on multiple tasks.BMC genomics · 2026Article
- Cytokines in cerebrospinal fluid combined with machine learning improve the diagnostic accuracy and predict the progression of neurosyphilis.Frontiers in immunology · 2026Article
- Editorial: Methods for imaging and omics data science: advances, applications, and spatiotemporal innovations.Frontiers in genetics · 2026Article
- Heimdall: A Modular Framework for Tokenization in Single-Cell Foundation Models.bioRxiv : the preprint server for biology · 2025Article
- Adaptive resampling for improved machine learning in imbalanced single-cell datasets.bioRxiv : the preprint server for biology · 2025Article
- KGG: Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions.Journal of chemical information and modeling · 2025Article
- Consequences of training data composition for deep learning models in single-cell biology.bioRxiv : the preprint server for biology · 2025Article
Corrections and comments
- Update of
Authors and funding
10 authors.
Funding
Abstract
The success of transformer-based foundation models on natural language and images has motivated their use in single-cell biology. Single-cell foundation models have been trained on increasingly larger transcriptomic datasets, scaling from initial studies with 1 million cells to newer atlases with over 100 million cells. Here we investigate the role of pretraining dataset size and diversity on the performance of single-cell foundation models on both zero-shot and fine-tuned tasks. Using a large corpus of 22.2 million cells, we pretrain a total of 400 models, which we evaluate by conducting 6,400 experiments. Our results show that current methods tend to plateau in performance with pretraining datasets that are only a fraction of the size of current training corpora. Unlike large language models, single-cell foundation models show no clear data scaling laws, indicating that developers should focus on balancing model capacity, dataset size and computational resources rather than indiscriminately increasing all three.
Indexed as
Identifiers
What Socratic holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the Socratic graph.