bio.rodeo
ModelsOrganizationsProvidersLeaderboardAboutSign in
bio.rodeo

The authoritative source for evaluating biological foundation models. No hype, just honest analysis.

Categories
  • DNA & Gene
  • RNA
  • Protein
  • Small molecule
  • Single-cell
  • Spatial omics
  • Pathology
  • Imaging
  • Metabolomics
  • Biosignals
  • Language model
bio.rodeoModelsOrganizationsProvidersLeaderboardAboutFAQSubmit a modelContact
© 2026 Pulsatance. All rights reserved. ~
Built by Pulsatance
models / dna-gene / segmentnt
DNA & Gene
InstaDeepBioNTechUniversity of CopenhagenReleased March 2024

SegmentNT

Genome annotation model that labels 14 genic and regulatory element classes at single-nucleotide resolution across DNA windows up to 50 kb.

The short version

  • —Locates exons, splice sites, promoters, and enhancers in a single pass over one window
  • —Treats annotation as instance segmentation, so overlapping elements are labeled at once
  • —A U-Net decoder upsamples token embeddings back to a prediction for every base
  • —Annotates genomes of species it never saw in training, including plants
15Openness
250HF downloads
906GitHub stars

Where to run it

No providers recorded yet. Browse all providers

Genome annotation — marking where genes, exons, splice sites, promoters, and enhancers sit along a chromosome — has long relied on a patchwork of specialist tools, each trained from scratch on one element class and a modest labeled dataset. That caps accuracy and transfers poorly to newly sequenced genomes, a growing problem as sequencing outpaces curation.

SegmentNT, built at InstaDeep with collaborators at BioNTech and the University of Copenhagen, replaces the patchwork with one model. It reframes annotation as instance segmentation, the task computer vision uses to label every pixel in an image, and applies it to DNA: every nucleotide in an input sequence receives a probability for each of 14 genic and regulatory element classes at once. Because the labels are not mutually exclusive, nested elements — an exon inside a protein-coding gene inside a promoter-proximal region — are handled natively.

The model attaches a one-dimensional U-Net decoder to a pretrained Nucleotide Transformer backbone, which supplies self-supervised representations learned from genomes across hundreds of species. The same framework was applied to two other DNA foundation models, Enformer and Borzoi, producing SegmentEnformer and SegmentBorzoi and showing the recipe is not specific to one encoder.

#Key Features

  • Fourteen element classes at once: Protein-coding genes, lncRNAs, 5′ and 3′ UTRs, exons, introns, splice donor and acceptor sites, poly(A) signals, tissue-invariant and tissue-specific promoters and enhancers, and CTCF-bound sites are all predicted in the same forward pass.
  • Single-nucleotide resolution: The U-Net decoder upsamples token-level embeddings back to per-base predictions, placing splice junctions and poly(A) signals at an exact position rather than inside a bin.
  • Long input windows: Trained on sequences up to 30 kb, the model retains its best accuracy on 50 kb inputs at inference — enough to hold most human genes end to end.
  • Cross-species transfer: A model trained only on human elements carries over to other animals, and a multispecies variant fine-tuned on human, mouse, chicken, zebrafish, fruit fly, and C. elegans reaches genomes never seen during training.
  • Encoder-agnostic framework: Swapping the backbone for Enformer or Borzoi pushes the input window to 196 kb and 524 kb respectively, trading some gene-level accuracy for stronger regulatory-element context.

#Technical Details

SegmentNT pairs the 500M-parameter nucleotide-transformer-v2-multi-species encoder with a 1D U-Net head of two downsampling and two upsampling convolutional blocks — roughly 53M additional parameters, 562M in total. Labels came from GENCODE V44 for gene elements and ENCODE's SCREEN registry for regulatory elements, with chromosomes 20 and 21 held out for testing and chromosome 22 for validation. Context was extended by curriculum: the 3 kb model saw 10.24 billion tokens, and successive checkpoints were retrained at 10 kb, 20 kb, and 30 kb.

Averaged over the 14 classes, SegmentNT-30kb reaches a Matthews correlation coefficient of 0.45, rising to 0.47 when inference runs on 50 kb sequences. Pretraining is what carries the result: at 3 kb context the model averages 0.37 against 0.16 for the identical architecture with a randomly initialized encoder, and 0.07 to 0.11 for one-hot U-Net baselines trained from scratch. On the SpliceAI test set it comes close to dedicated splice predictors, with auPRC of 0.93 for both donor and acceptor sites against 0.94 and 0.96 for SpliceAI and 0.94 for Pangolin on both. SegmentEnformer and SegmentBorzoi average 0.34 and 0.35 across the same elements. The multispecies variant averages 0.64 on animals close to human, 0.57 on distant animals, and 0.45 on plants.

#Applications

The most direct use is annotating newly sequenced genomes: a lab working on a non-model organism can obtain gene structures and regulatory elements without assembling a species-specific training set, and the multispecies model beat the HMM gene finder AUGUSTUS on gene annotation in every species tested except Arabidopsis. In human genomics, the per-nucleotide outputs support asking what a candidate sequence does — screening synthetic constructs for cryptic splice sites, checking whether a designed regulatory element carries promoter or enhancer signatures, or annotating regions where curated coverage is thin. Running from sequence alone, with no alignments or expression data, it fits in-silico design loops that must score many variants quickly.

#Impact

SegmentNT is among the clearest demonstrations that DNA foundation-model pretraining pays off on dense structured prediction rather than only on sequence-level classification: the gap between the pretrained and randomly initialized encoders is the paper's central result. The work also established segmentation as a reusable head for genomic backbones, with SegmentEnformer and SegmentBorzoi built by swapping the encoder alone. Two constraints bound practical use. The code and the two SegmentNT checkpoints — the human model and the multispecies model — are released under CC BY-NC-SA 4.0, which excludes commercial application, while the SegmentEnformer and SegmentBorzoi checkpoints carry no license tag at all on HuggingFace, leaving their reuse terms undefined. And the labels are human-derived, so element definitions carried into distant species inherit human annotation conventions — reflected in weaker scores on the diffuse classes, lncRNAs and CTCF-bound sites, even within human.

At a glance

Parameters
562 Million
Released
March 2024
Category
DNA & Gene
License
NOASSERTION
Organizations
InstaDeep / BioNTech / University of Copenhagen

Related models

  • Nucleotide Transformer v3 (NTv3)

    InstaDeep / Research Institute of Molecular Pathology (IMP) / Medical University of Vienna / Cornell University / Cold Spring Harbor Laboratory

  • Nucleotide Transformer

    InstaDeep

  • PlantGeneAnn

    Huazhong Agricultural University

  • DNABERT-Enhancer

    Stony Brook University / La Jolla Institute for Immunology / University of California, San Diego

  • GENERanno

    Alibaba Cloud / Zhejiang University / Institute of Dataspace

Links

GitHub RepositoryResearch PaperbioRxiv PreprintHuggingFace ModelDocumentationGoogle Colab

Tags

genome_annotationgenomicssegmentationtransfer_learningu_net

Something wrong?

Much of this page is generated or calculated automatically. Flag anything that looks off and we will re-run it.