A program to reconstruct plasmids using homology to plasmid clusters. Uses a similar approach as mob suite and uses the same replication database, but a slightly modified plasmid databases (with a few extra Pantoea agglomerans plasmids).
Only available on Linux systems.
Plasmid_recon.py
Main reconstruction program.
Build_plas_db.py
Builds the plasmid database for Plasmid_recon.
initialize_db.py
Downloads ORI and plasmid databases.
Performance_test.py
Tests performance using complete genomes. (Currently under work)
genome_frag.py
Randomly fragments genomes for testing.
utils.py
Functions to help abstract Plasmid_recon.py, Build_plas_db.py, Performance_test.py and download_db.py
- Clone this repository and enter directory:
cd Plasmid_reconstructor - Create environment and activate it:
conda env create -f environment.yaml -n plas
conda activate plas
- Begin by downloading the databases using the initialize_db.py script.
python initialize_db.py -t <your_num_threads> - Use Plasmid_recon.py on your fasta file.
python Plasmid_recon.py -i INPUT_FASTA -o OUTPUT_DIR -t NUM_OF_THREADS (optional)
Note: if you are not searching for Pantoea agglomerans plasmids, you need to provide a chromosome from your species of interest, and run
makeblastdb -dbtype nucl -in your_chromosome_fasta -out your_location
Use the --chromdb flags for usage of Plasmid_reconstructor.py
-
Create artificially fragmented genomes (all fragments from chromosome will be named chromosome, and plasmid with a number depending on the plasmid they are from).
python genome_frag.py -i Test_dataset -n NUM_FRAGS -o OUTDIR -
Run Plasmid_recon.py individually on fragmented genomes and see performance.
Note: An automated method to assess performance is being developed and will be available soon.
To see option available for any program, please run the following:
python Program_name.py -h