Saved in:
Bibliographic Details
Main Authors: Touron, Eloïse, Rodrigues, Pedro L. C., Arbel, Julyan, Varoquaux, Nelle, Arbel, Michael
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.21617
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918524549595136
author Touron, Eloïse
Rodrigues, Pedro L. C.
Arbel, Julyan
Varoquaux, Nelle
Arbel, Michael
author_facet Touron, Eloïse
Rodrigues, Pedro L. C.
Arbel, Julyan
Varoquaux, Nelle
Arbel, Michael
contents Inference from interaction maps, such as centromere identification from genome-wide chromosome conformation capture techniques -- notably Hi-C -- can be formulated as a generic inverse problem: infer a set of parameters given a map summarizing pairwise interactions between entities through blocks of variable numbers and sizes. In this work, we introduce a data-driven approach that leverages shared structure between these maps, such as global alignment between localized patterns, while handling the variability in number and size of entities 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 accurately recovers their genomic positions across a wide range of species of various genome sizes.
format Preprint
id arxiv_https___arxiv_org_abs_2605_21617
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle $\textit{BlockFormer}$ : Transformer-based inference from interaction maps
Touron, Eloïse
Rodrigues, Pedro L. C.
Arbel, Julyan
Varoquaux, Nelle
Arbel, Michael
Machine Learning
Quantitative Methods
Inference from interaction maps, such as centromere identification from genome-wide chromosome conformation capture techniques -- notably Hi-C -- can be formulated as a generic inverse problem: infer a set of parameters given a map summarizing pairwise interactions between entities through blocks of variable numbers and sizes. In this work, we introduce a data-driven approach that leverages shared structure between these maps, such as global alignment between localized patterns, while handling the variability in number and size of entities 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 accurately recovers their genomic positions across a wide range of species of various genome sizes.
title $\textit{BlockFormer}$ : Transformer-based inference from interaction maps
topic Machine Learning
Quantitative Methods
url https://arxiv.org/abs/2605.21617