genenet 0.0.4

Constructing gene association networks using chromosomal conformational capture technology based on three-dimensional space (3D-GeneNet)

3D-GeneNet

Constructing gene association networks using chromosomal conformational capture technology based on three-dimensional space (3D-GeneNet)

E-mail: [e-mail]

Introduction

3D-GeneNet is a tool for constructing gene association networks that use the chromosomal conformational capture data to estimate the spatial interaction frequency between genes and identify chromosomal interaction regions (CIDs). With normalized or unnormalized IF matrices and bacterial genome-wide Genbank annotation files as inputs, the software will construct the gene/protein interaction network, identify the CIDs, and export the network as a file.

Installation

Requirement:

• python (3.x)
• biopython
• numpy
• pandas
• scipy

Notes: You may alternatively install using the following conda commands shown below, which are derived from our built environment file environment.yaml.

conda env create -f environment.yaml -n new_env_name

How to use 3D-GeneNet?

Parameters

Parameter	Full name	Description	Default
-gb	genbank_file	Genbank filename with annotation information for the strain.	|
-p	multichromosome	Used to determine if the target strain is multichromosome; if the strain is multichromosome, set to 1 and is 0 by default.	0
-gbc	multichromosomeGenbank	If the strain is multichromosome, enter the Genbank filenames for the chromosomes in order, separated by a '/' symbol.	|
-if	interaction_file	Contact matrix filename to convert interaction frequencies between bins to interaction frequencies between genes.	|
-r	resolution	Resolution of interaction_file.	|
-i	input_file_path	Path to the input file directory.	|
-b	bin_begin_number	Begin the bin serial number in the contact matrix, which is 1 by default.	1
-n	n_bootstrap	Number of bootstrap iterations,which is None by default.	None
-s	seed	Seed for random number generation to ensure reproducibility.	None
-q	quantile_threshold	This parameter selects the percentile of p-values, with the default value set at 95. This means it filters for the top 5% of gene pairs based on their p-values.	95
-d	remove_distance	Linear genomic distances to select removed gene pairs and is 0 by default.	0
-ic	cid_interaction_file	Contact matrix filename to identify CIDs and is interaction_file by default. If cid_interaction_file = None, then it will not identify CIDs	|
-rc	resolution_cid	Resolution of CIDmatrix.	|
-o	outfile_path	Output folder name.	output
-h	help	Help documentation	|

Usage

python 3D-GeneNet.py -i [input_file_path] -gb [genbank_file] -if [interaction_file] -ir [resolution] [...]

For example

• Single-chromosome The GSM2870409 sample of Escherichia coli K-12 MG1655 was taken is an example to demonstrate the complete function.

# Run 3D-GeneNet
python 3D-GeneNet.py -i "D:\Data\Project\3D-GeneNet\software_CLI\3D-GeneNet\input\ecoli" -gb U00096.3.gb -if rawdata_1000_iced.matrix -ic rawdata_10000_iced.matrix -o "D:\Data\Project\3D-GeneNet\software_CLI\3D-GeneNet\output\ecoli" -r 1000 -rc 10000 -q 99.5
------------------------------
2024-05-08 15:50:55
Run 3D-GeneNet

------------------------------
2024-05-08 15:50:55
Convert interaction frequency

------------------------------
2024-05-08 15:51:52
Select gene pairs

------------------------------
2024-05-08 16:44:55
Identify CIDs

------------------------------
2024-05-08 16:44:58
Output files

Time:    3247.2316389083862s

• Multi-chromosome The SRR3180951 sample of Vibrio cholerae was taken is an example to demonstrate the complete function.

#Run 3D-GeneNet
python 3D-GeneNet.py -i "D:\Data\Project\3D-GeneNet\software_CLI\3D-GeneNet\input\vibrio_cholerae" -p 1 -gbc vibrio_cholerae_1.gb/vibrio_cholerae_2.gb -if SRR3180951_1000_iced.matrix -ic SRR3180951_10000_iced.matrix -o "D:\Data\Project\3D-GeneNet\software_CLI\3D-GeneNet\output\vibrio_cholerae" -r 1000 -rc 10000 -q 99.5
------------------------------
2024-05-08 17:10:40
Run 3D-GeneNet

------------------------------
2024-05-08 17:10:40
Convert interaction frequency

------------------------------
2024-05-08 17:11:13
Select gene pairs

------------------------------
2024-05-08 17:21:01
Identify CIDs

------------------------------
2024-05-08 17:21:05
Output files

Time:    626.8358745574951s

Detailed usage

1. -i input_file_path Put all input files into the 'input_file_path', folder before starting the run. 'input' is the folder name where all input files are located.
2. -o outfile_path 'outfile_path' is the name of the folder where all output files are located. The software will output 5 files if it identifies CIDs. 'cid_t_pvalue.csv' and 'cid_bin_preference.pdf' are files describing bins' t and pvalue. 鈥檆id_number.csv' is the range of CIDs. 'gene_association_network.txt' and 'gene_cid.txt' are files from the obtained gene association network and the CID attribute of the gene. 'gene_association_network.txt' is visualized by Cytoscape. Its specific structure is as follows:

   + input
   	++ U00096.3.gb (genbank_file)
   	++ GSM2870409_1000_iced.matrix (interaction_file)
   	++ GSM2870409_10000_iced.matrix (cid_interaction_file)
   + output
   	++ cid_bin_preference.pdf
   	++ cid_number.csv
   	++ cid_t_pvalue.csv
   	++ gene_cid.csv
   	++ gene_association_network.csv

3. -gb genbank_file This command inputs the Genbank annotation file of the strain, which needs to contain the genome length and the position information of the gene, and its format is as follows:

LOCUS       U00096               4641652 bp    DNA     circular BCT 23-SEP-2020
DEFINITION  Escherichia coli str. K-12 substr. MG1655, complete genome.
           ......
           ......
    gene            190..255
                    /gene="thrL"
                    /locus_tag="b0001"
                    /gene_synonym="ECK0001"
                    /db_xref="ASAP:ABE-0000006"
                    /db_xref="ECOCYC:EG11277"
    CDS             190..255
                    /gene="thrL"
                    /locus_tag="b0001"
                    /gene_synonym="ECK0001"
                    /codon_start=1
                    /transl_table=11
                    /product="thr operon leader peptide"
                    /protein_id="AAC73112.1"
                    /db_xref="UniProtKB/Swiss-Prot:P0AD86"
                    /db_xref="ASAP:ABE-0000006"
                    /db_xref="ECOCYC:EG11277"
                    /translation="MKRISTTITTTITITTGNGAG"

4. -im interaction_file -r resolution This command inputs the contact matrix file that converts the interaction frequency is three column data. The resolution is this matrix resolution, 1000 is recommended. Three columns format The data consists of three columns: the first column is the serial number of the first bin, the second column is the serial number of the second bin, and the third column is the interaction frequency of the two bins.

   0	0	65.478946
   0	1	633.672990
   0	2	9.645624
   0	3	5.717007
   0	4	5.582881
   ...	...	...
   4641	4641	532.456055
   

5. -ic --cid_interaction_file -rc resolution_cid If you need to identify CIDs, provide the contact matrix in the same format as the above interaction_file; 10000 is the suggested resolution. If you do not need to identify CIDs, enter the command -ic None.
6. -p --multichromosome -gbc --multichromosomeGenbank If the strain is multi-chromosome, set -p 1. Genbank filenames for chromosomes are entered in order after -gbc, separated by a '/'.