microbiONT: an AI-assisted, privacy-focused platform for local Nanopore 16S and 18S amplicon analysis
Che-Chun Chen, Hsin-Yun Lu, Ying-Ning HoAbstract
Motivation
The affordability and high-throughput capabilities of Oxford Nanopore Technologies (ONT) have democratized genomic sequencing, yet the complexity of bioinformatics environment setup and analysis remains a barrier for many biologists.
Results
microbiONT addresses this gap by providing a user-friendly, privacy-focused platform that features a streamlined, one-command installation process. Integrating a local Large Language Model (Llama 3.1) with an intuitive graphical interface, microbiONT serves not only as an analysis tool but also as an AI copilot to assist users in learning bioinformatics concepts. By translating natural language requests into executable commands and simplifying deployment, microbiONT empowers non-experts to perform rigorous or custom 16S/18S amplicon analysis locally. This ensures rapid data insights without the need for command line use. Beyond its AI-integrated GUI, the novelty of microbiONT lies in its fully localized, highly accurate, and privacy-preserving architecture for flexible data analysis.
Availability and implementation
Source code and portable binaries are freely available at https://github.com/MBEBlab/microbiONT under the MIT license.