Abdul Muntakim Rafi - "Detecting and avoiding homology-based data leakage in genome-trained sequence models" (Rafi et al.)

17 Mar 2026

Overview

Paper Information

• Title: Detecting and avoiding homology-based data leakage in genome-trained sequence models.
• Authors: Abdul Muntakim Rafi, Brett Kiyota, Nozomu Yachie, Carl de Boer
• Journal: bioRxiv
• DOI & Link: https://doi.org/10.1101/2025.01.22.634321

Contact

For questions or further discussion, please reach out to genomicdeeplearning@cs.jhu.edu