Learning Genomic Sequence Representations using Graph Neural Networks over De Bruijn Graphs

Kacper Kapuśniak, Manuel Burger, Gunnar Rätsch, Amir Joudaki
12/6/2023
cs.LGq-bio.GN

No official code found. Best implementation: ratschlab/genomic-gnn (7 stars, MIT, last commit 2024-11-11).

Abstract

The rapid expansion of genomic sequence data calls for new methods to achieve robust sequence representations. Existing techniques often neglect intricate structural details, emphasizing mainly contextual information. To address this, we developed k-mer embeddings that merge contextual and structural string information by enhancing De Bruijn graphs with structural similarity connections. Subsequently, we crafted a self-supervised method based on Contrastive Learning that employs a heterogeneous Graph Convolutional Network encoder and constructs positive pairs based on node similarities. Our embeddings consistently outperform prior techniques for Edit Distance Approximation and Closest String Retrieval tasks.

View on arXivView PDF

Code Implementations(1)

ratschlab/genomic-gnnCommunity100%
MIT

Repository for the paper "Learning Genomic Sequence Representations using Graph Neural Networks over De Bruijn Graph"

73PythonDec 2, 20231 years agoMIT

Similar Papers

Vector Embeddings by Sequence Similarity and Context for Improved Compression, Similarity Search, Clustering, Organization, and Manipulation of cDNA Libraries

85% match2023

An Independent Implementation of Quantum Machine Learning Algorithms in Qiskit for Genomic Data

83% match2024

HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences

82% match2026

DNAZEN: Enhanced Gene Sequence Representations via Mixed Granularities of Coding Units

82% match1 repo32025

Associative Knowledge Graphs for Efficient Sequence Storage and Retrieval

82% match1 repo2024

Context-Aware Regularization with Markovian Integration for Attention-Based Nucleotide Analysis

81% match1 repo82025

Cite this paper

@article{kapuniak2023learning,
  title  = {Learning Genomic Sequence Representations using Graph Neural Networks over De Bruijn Graphs},
  author = {Kacper Kapuśniak and Manuel Burger and Gunnar Rätsch and Amir Joudaki},
  year   = {2023},
  eprint = {2312.03865},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2312.03865v1}
}

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/2312.03865.svg)](https://codeofpaper.com/paper/2312.03865)

Discussion