DeepInverton

Introduction

The DeepInverton algorithm is designed to detect DNA inversion mediated phase variation in bacterial genomes just using nucleotide sequence. It works by identifying regions flanked by inverted repeats and identifying regions using the DeepInverton model. DeepInverton should be valuable to the research community, as it enables researchers to effectively identify a list of candidate invertons from massive data, provide good targets for further exploration of invertons.

Installation

We recommend deploying DeepInverton using conda

# clone this repository
git clone https://github.com/HUST-NingKang-Lab/DeepInverton.git
cd DeepInverton
# configure environment using environment.yml (Time consumption of ~6s)
conda env create -f environment.yml
# activate the environment
conda activate deepinverton

Usage

Perform a search and identification of nucleotide sequences, including assembled contigs, genomics or single sequences.

python deepinverton.py -f input_sequence.fna -o result_dir_path --model /deepinverton/model/DeepInverton.pth -x prefix_filename  -g 15 85 -p

Quick Start

All you need to get started just are nucleotide sequences (in fasta format). Then, you can search for the invertons using DeepInverton.

To test DeepInverton, you can use the example files (genomic.txt)

example:

python deepinverton.py -f /deepinverton/example/genomic.fna -o /deepinverton/example/result --model /deepinverton/model/DeepInverton.pth -x genomic  -g 15 85 -p

If successful, the output will be in /deepinverton/expample/result/ with three files, including genomic_ir.txt, genomic_inverton.txt and genomic_ir_possibility.txt.(Time consumption of ~4min)

Parameters deepInverton.py

• -f --fasta input nucleotide sequence file in fasta format
• -o --outdir where output files should be written. default is current working directory.
• -x --prefix base name for output files
• -d --model the path of DeepInverton model
• -e --einv einverted parameters, if unspecified run with DeepInverton default pipeline
• -m --mismatch max number of mismatches allowed between IR pairs, used with -einv (default:3)
• -r --IRsize max size of the inverted repeats, used with -einv (default:50)
• -g --gcrange the minimum and maximum value of GC ratio
• -p --polymer Eliminate homopolymer inverted repeats

Result file

The DeepInverton program will generate three output files, including prefix_ir.txt, prefix_inverton.txt and prefix_ir_possibility.txt.

filename	description
prefix_ir.txt	output table with inverted repeats coordinates
prefix_inverton.txt	output table with invertons coordinates
prefix_ir_possibility.txt	output table with inverted repeats possibility of invertons

• prefix_ir.txt, prefix_inverton.txt

Column name	Explanation
ID	The sequence name of inverted repeat combined with Scaffold, pos A, pos B, pos C and pos D
Scaffold	The sequence name where the inverted repeat is detected
PosA	The start coordinate of the first inverted repeat (0-based)
PosB	The end coordinate of the first inverted repeat (1-based)
PosC	The start coordinate of the second inverted repeat (0-based)
PosD	The end coordinate of the second inverted repeat (1-based)
IrA	The sequence of the first inverted repeat
Mid	The sequence of the invertible promoter
IrB	The sequence of the second inverted repeat

• prefix_ir_possibility.txt

Column name	Explanation
ID	The sequence name of inverted repeat combined with Scaffold, pos A, pos B, pos C and pos D
positive	The inverton possibility of the inverted repeat
negative	The non-inverton possibility of the inverted repeat