Genomic Interpreter: A Hierarchical Genomic Deep Neural Network with 1D Shifted Window Transformer

Zehui Li, Akashaditya Das, William A V Beardall, Yiren Zhao, Guy-Bart Stan
6/8/2023
cs.LGq-bio.GN

No official code found. Best implementation: Zehui127/1d-swin (23 stars, MIT, last commit 2023-07-03).

Abstract

Given the increasing volume and quality of genomics data, extracting new insights requires interpretable machine-learning models. This work presents Genomic Interpreter: a novel architecture for genomic assay prediction. This model outperforms the state-of-the-art models for genomic assay prediction tasks. Our model can identify hierarchical dependencies in genomic sites. This is achieved through the integration of 1D-Swin, a novel Transformer-based block designed by us for modelling long-range hierarchical data. Evaluated on a dataset containing 38,171 DNA segments of 17K base pairs, Genomic Interpreter demonstrates superior performance in chromatin accessibility and gene expression prediction and unmasks the underlying `syntax' of gene regulation.

View on arXivView PDF

Code Implementations(1)

Zehui127/1d-swinCommunity100%
MIT

The implementation of 1d-swin, an efficient transformer for capturing hierarchical 1-dimentional long range sequence

235PythonJun 26, 20233 years agoMIT

Similar Papers

Cite this paper

@article{li2023genomic,
  title  = {Genomic Interpreter: A Hierarchical Genomic Deep Neural Network with 1D Shifted Window Transformer},
  author = {Zehui Li and Akashaditya Das and William A V Beardall and Yiren Zhao and Guy-Bart Stan},
  year   = {2023},
  eprint = {2306.05143},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2306.05143v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2306.05143.svg)](https://codeofpaper.com/paper/2306.05143)

Discussion