Good News: We have built a Galaxy server at http://hts.iit.edu/galaxy. You can go to http://hts.iit.edu/galaxy/tool_runner?tool_id=plasmidhunter to analyze your data by simply uploading it and waiting.
presented by Institute of Food Safety and Health, Illinois Institute of Technology
Plasmids are extrachromosomal DNA found in microorganisms. They often carry beneficial genes that help bacteria adapt to harsh conditions. Plasmids are also important tools in genetic engineering, gene therapy, and drug production. However, it can be difficult to identify plasmid sequences from chromosomal sequences in genomic and metagenomic data. Here, we have developed a new tool called PlasmidHunter, which uses machine learning to predict plasmid sequences based on gene content profile. PlasmidHunter achieved high accuracies (up to 97.6%) and high speeds in benchmark tests including both simulated contigs and real metagenomic plasmidome data, outperforming other existing tools.
Keywords: artificial intelligence (AI), machine learning (ML), plasmid prediction, genomic sequencing
conda create -n plasmidhunter python=3.10
conda activate plasmidhunter
conda install -c conda-forge -c bioconda -y diamond=2.1.8 prodigal # Or use mamba install. For Windows users, please install the two packages manually instead.
pip install plasmidhunter
plasmidhunter -h
The result is a tab-delimited table showing the prediction of each sequence. The columns include Prediction (0: chromosome, 1: plasmid), Probability of 0 (chromosome), and Probability of 1 (plasmid).
v1.1 9/1/2022
PlasmidHunter is now using much less memory.
v1.2 11/23/2023
PlasmidHunter has an expanded database for a higher annotation rate.
PlasmidHunter is now accepting shorter contigs down to 1 Kbp and has a higher accuracy for short contigs.
v1.3 4/2024 Fixed some minor bugs
v1.4 5/17/2024
Converted model and feature pickle file into parameter file and text file, respectively.
PlasmidHunter: Accurate and fast prediction of plasmid sequences using gene content profile and machine learning
Renmao Tian, Jizhong Zhou, Behzad Imanian
bioRxiv 2023.02.01.526640; doi: https://doi.org/10.1101/2023.02.01.526640
If you have any questions, please contact Renmao Tian (tianrenmao[at]gmail.com) or Behzad Imanian (bimanian[at]iit.edu).
Educational Community License, Version 2.0