Contents:

• Quick Start
• Installation and Usage
• Output Format
• Visualisation and Analysis
• Examples
• Troubleshooting
• Release Notes

Overview

ORILINX is a large language model built on DNABERT-2 to detect the location of replication origins in mammalian genomes.

ORILINX is under active development by the Boemo Group based in the Department of Pathology, University of Cambridge. We aim to push regular updates and improvements and incorporating new functionality is an active area of our computational research.

Publications

If you use ORILINX for your research, please cite our publication:

Pfuderer PL, Berkemeier F, Nassar J, Crisp A, Jaworski JJ, Moore J, Sale JE, Boemo MA. A genome language model for mapping DNA replication origins. bioRxiv 2026. [Read on bioRxiv]

Bugs, Questions, and Comments

Should any bugs arise or if you have any questions about usage, please raise a GitHub issue. Your feedback is an important part of the development process. If you have comments or suggestions to improve the software or the documentation, please Email Michael Boemo at mb915@cam.ac.uk.