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 / fiberfold
DNA & Gene
Stanford UniversityUniversity of WashingtonUniversity of OxfordBrotman Baty Institute for Precision MedicineChan Zuckerberg BiohubReleased February 2025

FiberFold

Predicts haplotype-specific 3D genome organization and Hi-C contact maps from a single long-read Fiber-seq assay, using no DNA sequence as input.

The short version

  • —Resolves 3D genome folding separately for each parental haplotype from a single assay
  • —Uses chromatin accessibility, CpG methylation, and CTCF footprints in place of DNA sequence
  • —Predicts TAD changes around a patient translocation from a donor-specific assembly
  • —Reveals attenuated TADs on the inactive X chromosome in human cells
13Openness2Citations
13GitHub stars

Where to run it

No providers recorded yet. Browse all providers

Hi-C and Micro-C measure how the genome folds, but they average over both copies of every chromosome. Fewer than 10% of Hi-C reads can be assigned to a haplotype, so the two parental copies of a locus — which may differ in accessibility, methylation, or structure — collapse into one contact map. That matters most at structural variants, in repetitive regions, and on the X chromosome, where the active and inactive copies fold differently but have been studied almost entirely in mice.

FiberFold, from the labs of Nicolas Altemose at Stanford University and Andrew Stergachis at the University of Washington, takes a different route. Fiber-seq, a PacBio long-read assay, reports chromatin accessibility, CpG methylation, and CTCF occupancy on individual DNA molecules long enough to phase. FiberFold maps those signals — plus CTCF motif orientation — onto a Hi-C contact map at 10 kb resolution across a roughly 2 Mb window, so one sequencing run yields accessibility, methylation, protein occupancy, genetic variation, and 3D structure together.

The model deliberately excludes DNA sequence. Sequence-to-structure genome models generalize poorly across cell types; conditioning only on measured epigenomic state leaves FiberFold free to predict differently for two haplotypes or two cell types sharing a reference sequence.

#Key Features

  • One assay in, contact map out: Chromatin accessibility (Fiber-seq Inferred Regulatory Elements), CpG methylation, CTCF footprint scores, and CTCF motif direction all come from the same Fiber-seq library, replacing an ATAC-seq plus ChIP-seq plus Hi-C stack.
  • Sequence-free by construction: Dropping the DNA encoder halves the initial encoder block's parameter count and forces the model to condition on measured chromatin state rather than the reference genome.
  • Haplotype-resolved predictions: Because long reads phase cleanly, the model runs separately on each parental haplotype, producing two contact maps where a Hi-C experiment produces one.
  • Cross-cell-type transfer: Trained only on the lymphoblastoid line GM12878, it recovers the differential contacts that distinguish the leukemia line K562.
  • Works on rearranged genomes: For a patient-derived line carrying a balanced X;13 translocation, predictions ran against a donor-specific phased assembly rather than the reference, modeling the derivative chromosomes as they actually exist.

#Technical Details

The architecture adapts C.Origami: an input window of 2,097,152 bp passes through 13 one-dimensional residual convolution blocks compressing it to an 8,192 bp latent resolution, then 8 transformer blocks capturing long-range relationships, then 5 two-dimensional dilated residual convolution blocks that emit the 10 kb contact map. Hi-C targets are log-transformed and input tracks depth-normalized to a 0-1 range; training minimizes mean squared error against experimental maps. Data came from GM12878 Fiber-seq paired with published Hi-C, tiled as 2 Mb windows on a 36 kb sliding step with ENCODE blacklist regions and coverage gaps removed; augmentation applied a random shift of up to 10 kb per epoch, Gaussian noise, and a 50% chance of reversing both matrices. Chromosome 10 was held out for validation and chromosome 15 for testing, on 4 A100 GPUs with Adam at a learning rate of 0.0004 and a batch size of 8.

On held-out chr15 the model reached a median map-to-map Pearson r of 0.938 and Spearman r of 0.906, pixel-wise Pearson r of 0.932, and a median insulation correlation of 0.836 at a 500 kb radius. Transferred to K562 without retraining, map-to-map correlations held at 0.930 Pearson and 0.902 Spearman, with pixel-wise r of 0.871. Haplotype-averaged predictions matched experimental Hi-C at r = 0.938 on autosomes and 0.91 on chrX, and the per-haplotype maps showed the inactive maternal X losing sub-megabase TADs and distal contacts relative to the active paternal copy — attenuation, not the near-complete TAD loss reported in mice.

#Applications

The immediate use is rare-disease genomics. For patient UDN318336, carrying a 46,XX,t(X;13)(p22.1;q14.1) translocation, phased contigs spanning both breakpoints yielded a sawtooth pattern of TAD differences between the derivative and intact X chromosomes, showing how a structural variant reorganizes local chromatin. More broadly, any laboratory already generating Fiber-seq data can add 3D structure to its readout without a separate Hi-C experiment — useful for variant prioritization, allele-specific regulation, and regions short reads cannot map.

#Impact

FiberFold is a bioRxiv preprint. Its central claim — that accessibility, CTCF binding, and CpG methylation alone suffice to reconstruct large-scale chromatin structure at accuracies comparable to sequence-based models — reframes what the sequence encoder in this model class contributes, and its human X-inactivation result shows a question phased predictions can answer and bulk Hi-C cannot. Training and prediction code is on GitHub, and the pretrained checkpoint — a 172 MB file in a lab-hosted S3 bucket — is retrievable from the bucket's object endpoint, though the wget command in the README points at an AWS console URL and fails. The repository carries no license file, so reuse terms are undefined. The model is fixed to one window size and one resolution and requires Fiber-seq input, tying adoption to PacBio workflows; the authors point to Micro-C training data as the route past the first constraint.

At a glance

Released
February 2025
Category
DNA & Gene
Organizations
Stanford University / University of Washington / University of Oxford / Brotman Baty Institute for Precision Medicine / Chan Zuckerberg Biohub

Related models

  • MIX-HIC

    Hong Kong University of Science and Technology (Guangzhou) / South China Normal University / City University of Hong Kong / Hong Kong University of Science and Technology

  • TRUHiC

    Temple University / University of Southern California / University of Washington / Howard Hughes Medical Institute

  • UniversalEPI

    ETH Zurich / SIB Swiss Institute of Bioinformatics / Swiss Data Science Center / EPFL / Dana-Farber Cancer Institute / Broad Institute / Harvard University / Helmholtz Munich / Technical University of Munich

  • ORIGAMI

    Zhejiang University / Hangzhou City University / Shanghai Jiao Tong University / City University of Hong Kong / Iowa State University

  • Evo2HiC

    University of Washington

  • ARCH3D

    University of Michigan

Links

GitHub RepositorybioRxiv Preprint

Tags

chromatinchromatin_contact_predictionchromatin_organizationcnngenomicsmultimodaltransformer

Something wrong?

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