BlockFormer : Transformer-based inference from genomic contact maps
Abstract
Locating genomic features from genomic contact maps, such as centromere identification from genome-wide chromosome conformation capture techniques, notably Hi-C, can be formulated as an inverse problem: infer one parameter per entity given a map summarizing pairwise interactions through blocks of variable numbers and sizes. In this work, we introduce a data-driven approach that leverages shared structure between these contact maps, such as global alignment between localized patterns, while handling the variability in number and size of chromosomes arising in real-world data. Our approach relies on a transformer architecture capable of handling such variability and a custom simulator to generate abundant, yet computationally cheap synthetic data for training. Applied to the problem of centromere localization, the method recovers genomic positions at or below the map resolution across species with various genome sizes using a coarse-to-fine refinement step when chromosome sizes fall outside the training range. We concentrate our evaluation on Hi-C data, where the block structure is well characterized and ground truth is available.
Related
- DiffDomain-Spectrum identifies structurally reorganized TADs from sparse aggregated single-cell Hi-C contact maps
- CREST: A Cortical Resting-State EEG Spatial Transformer for Chronic Pain Inference
- Two blind spots in the demographic inference of human origins from genomic data
- Robust Demographic Inference from Genomic and SNP Data
- CipherGenome: Homomorphic Inference for Genomic Mixture-of-Experts
BibTeX
@misc{touron2026blockformer,
title = {{$\textitBlockFormer$ : Transformer-based inference from genomic contact maps}},
author = {Eloïse Touron and Pedro L. C. Rodrigues and Julyan Arbel and Nelle Varoquaux and Michael Arbel},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.21617},
eprint = {2605.21617},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.21617},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - BlockFormer : Transformer-based inference from genomic contact maps
AU - Eloïse Touron
AU - Pedro L. C. Rodrigues
AU - Julyan Arbel
AU - Nelle Varoquaux
AU - Michael Arbel
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2605.21617
UR - https://arxiv.org/abs/2605.21617
AB - Locating genomic features from genomic contact maps, such as centromere identification from genome-wide chromosome conformation capture techniques, notably Hi-C, can be formulated as an inverse problem: infer one parameter per entity given a map summarizing pairwise interactions through blocks of variable numbers and sizes. In this work, we introduce a data-driven approach that leverages shared structure between these contact maps, such as global alignment between localized patterns, while handling the variability in number and size of chromosomes arising in real-world data. Our approach relies on a transformer architecture capable of handling such variability and a custom simulator to generate abundant, yet computationally cheap synthetic data for training. Applied to the problem of centromere localization, the method recovers genomic positions at or below the map resolution across species with various genome sizes using a coarse-to-fine refinement step when chromosome sizes fall outside the training range. We concentrate our evaluation on Hi-C data, where the block structure is well characterized and ground truth is available.
ER -