PathoLM: Identifying Pathogenicity from the DNA Sequence Through the Genome Foundation Model
Published:
PathoLM adapts pretrained DNA foundation models for pathogenicity identification from bacterial and viral sequences. It targets data-efficient detection of novel and divergent pathogens and evaluates zero-shot, few-shot, and species-classification settings.
@inproceedings{dip2024patholm,
title = {PathoLM: Identifying Pathogenicity from the DNA Sequence Through the Genome Foundation Model},
author = {Dip, Sajib Acharjee and Shuvo, Uddip Acharjee and Chau, Tran and Song, Haoqiu and Choi, Petra and Wang, Xuan and Zhang, Liqing},
booktitle = {Proceedings of the 19th Machine Learning in Computational Biology Meeting},
series = {Proceedings of Machine Learning Research},
volume = {261},
pages = {153--161},
year = {2024},
publisher = {PMLR}
}