What it is
The Nucleotide Transformer family are large language models pretrained on thousands of genomes that transfer to many genomic prediction tasks with minimal fine-tuning. Built by InstaDeep (acquired by BioNTech), they helped establish self-supervised pretraining as a genomics workhorse.
Evidence trail
BioAtlas keeps the path from source to decision visible. A connection records provenance; it does not imply that evidence is sufficient for every context.
Model passport
How Nucleotide Transformer represents biology
Category is navigation. These fields describe the model-specific computational transformation and deliberately override broad category defaults.
Biological scale
Modalities & tasks
Registry, claims and frontier intelligence
Version history not yet curated
1 version record · release year not yet normalized. Model-family identity remains separate from capability and access changes.
Explore version lineage →0 normalized claims
No task, dataset, split and metric claim has been normalized for this record yet.
Open claim intelligence →1 connected frontier
Regulatory genome · Peer-reviewed capability
Inspect research horizon →Connected research frontiers
These records describe active research directions, not guaranteed capabilities of this model. Evidence stages and unresolved questions are preserved separately.
Million-base regulatory variant prediction
Google DeepMind · 2026-01-28Can a single model predict how coding and non-coding variants alter expression, splicing, chromatin and regulatory binding over long genomic context?
Evidence boundary and unresolved questions
The model is a research predictor, not a personal-genome or clinical diagnostic system; tissue specificity and very long-range enhancer logic remain limitations.
- How reliably do predictions transfer to rare cell states and patient contexts?
- Can causal mechanisms be separated from learned correlations?
- How should predictions be prospectively validated?
regulatory genomics · variant effects · non-coding DNA · splicingOpen frontier record →Inputs and outputs
Inputs
DNA sequenceOutputs
Sequence predictionsEmbeddings or generated sequenceScientific and technical profile
Scientific principles
Technology
Scientific lineage
These are transparent concept matches—not claims that one scientist alone caused this model. Each connection is based on the model’s recorded domain, scientific principles, technical terms or an explicit lineage link.
DNA as the hereditary transforming principle
Oswald Avery, Colin MacLeod & Maclyn McCartyGenomics, variant interpretation, gene therapy and sequence foundation models depend on DNA being the durable molecular carrier of biological information.
The DNA double helix and complementary base pairing
James Watson & Francis CrickSequence analysis, variant prediction, genome design and nucleic-acid therapeutics all rest on this structural logic.
Reading the sequences of proteins and DNA
Frederick SangerBiological foundation models exist because proteins and genomes became readable, comparable and computable at scale.
Transformer self-attention
Ashish Vaswani and colleaguesProtein, genome, molecule and single-cell foundation models use attention to learn dependencies across biological sequences and multimodal inputs.
Information, entropy and communication
Claude E. ShannonSequence modelling, cross-entropy training, language models, mutual information and representation learning all use Shannon’s framework.
X-ray evidence for the helical structure of DNA
Rosalind Franklin & Raymond GoslingStructural genomics and sequence-to-structure reasoning began with experimentally grounded molecular geometry.
Evaluation evidence
BioAtlas has not yet extracted a structured benchmark claim for this record.
Known limitations
- Performance depends on the evaluation dataset and operating conditions.
- A structured benchmark claim has not yet been extracted for this record.
- Outputs require task-specific scientific and experimental validation.
Milestones
InstaDeep was acquired by BioNTech.
Open weights across several model sizes.