Plaspline, a pipeline for plasmid analysis form metagenome

Introduction

Plaspline is a snakemake based workflow, which aims to get a comprehensive analysis of plasmid, both in gene and plasmid community level from short-gun sequences. The main steps are quality control, assembly, circular/linear plasmids isolation, circular/linear plasmids genome verification and classification, and plasmids-relative-genes analysis.

Requirements

ruamel.yaml=0.16.12
click=7.1.2
snakemake=7.22.0
biopython=1.78
python=3.9

Install

create a conda environment and insatll dependencies:

> conda create -n plaspline -c conda-forge -c bioconda python=3.9 biopython=1.78 snakemake=7.22.0 click=7.1.2 ruamel.yaml=0.16.12

Clone package:

> git clone https://github.com/Wanli-HE/Plaspline.git 

all scripts are under the 'Plaspline' folder

Downlanding dependent databases and tools:

*** make sure that you have enough space in your folder for it (80G maybe!!!!)

downloading the databases and tools

> python Plaspline/bin/run_main.py downloading       

*** Notice: when downloading plasforest database, it requires input an e-mail address in your terminal, to get full database or input "no"!!!

Usage

Prepare config.yaml and sample.json file.

> python  Plaspline/bin/run_main.py preprocessing --path Plaspline/example/reads --adapter Plaspline/example/adapters.fa.gz --phix Plaspline/example/phix174_ill.ref.fa.gz

Notice:

1. In your reads folder: plaspline can only recognize the raw reads (forward and resvers) file which suffix is "_R1" or "_r1" or "_1" (only show forward in example).
   ***must be gzip file. otherwise in "quality control step" it will raise error.

2. If you want not to do the quality control step, but still have to give the "--adapter" and "--phix" input, (a fake path is ok, for example ./ ). don`t forget change "--skip_qc " to "True", which means jump quality control step.

3. the assembler defult is spades, you can also choose megahit as assembler, by input the params "--assembler megahit".

4. in command line "Plaspline/bin/run_main.py" is relative path, if you are not running the command with same folder of Plaspline, you need to input absolute path (abs.path/Plaspline/bin/run_main.py)

for more information, "python ~Plaspline/bin/run_main.py preprocessing -h"

Working

Dry running

Before true runing plaspline, it is better to check whether there is any command line errors, by dry run (" -n ").

  > python Plaspline/bin/run_main.py working  <"step"> -j 3  -n 

Notice:

1. before running, make sure the "cofig.yaml" and "samples.json" are under the floder which you are running the command.
2. **** choosing jobs (-j):
   • in config.yaml file, it records {threads} which is each rules threads, so in "-j (pnum)", this pnum <= sample_number * {threads}
   • running in cluster, -j {pnum}; this pnum <= sample number

True running:

> python Plaspline/bin/run_main.py working  "step" -j 3 

a. Runing plaspline all:

*** before running it, you have to make sure that scapp_k == max ker (that you are using in assembly step) in config.yaml file

> python Plaspline/bin/run_main.py working  "all" -j 3

b. Runing plaspline in each steps: (suggestion!!!)

1.quality control step

> python Plaspline/bin/run_main.py working "qc"  -j 3

2. assembly step

> python Plaspline/bin/run_main.py working "assembly"  -j 3

3. circular step

before running this step, check the max k-mer you are using in assembly step, and modifying scapp_k to "max k-mer (assembly)" in config.yaml file.

> python Plaspline/bin/run_main.py working "circilar"  -j 3

4. isolation linear plasmid contigs from all assembled contigs

> python Plaspline/bin/run_main.py working "isolation"  -j 3

5. non-redundant circular plasmid contig set step

> python Plaspline/bin/run_main.py working "contig_circular"  -j 3

6. contig circular classify step

> python Plaspline/bin/run_main.py working "contig_circular_classify"  -j 3

7. non-redundantlinear plasmid contig set step

> python Plaspline/bin/run_main.py working "contig_linear"  -j 3

8. circular plasmid gene analysis step

> python Plaspline/bin/run_main.py working "gene_circular"  -j 3

9. linear plasmid gene analysis step

> python Plaspline/bin/run_main.py working "gene_linear"  -j 3

Output:

1. qc_reads:

    "*.fastq.gz"            ----  quality control step results

2. assembly_res

    "*_assembly_res"        ----  assembly results 

3. circular:

    "chromosome/*_verify_chromosome_circular.fasta"  ----     chromosome contigs identified by plaspline
    "plasmid/*_verify_plasmid_circular.fasta"        ----     circular plasmid contigs identified by plaspline  
    "*_scapp_res"                                    ----     circular plasmid Contigs identified by scapp 
    "*_metaplasmidspades"                            ----     circular plasmid Contigs identified by metaplasmidspades

4. circular_non_redundant_contig:

    "circular_non_redundant_contigs_rep_seq.fasta"             --- circular plasmid catalog, 
    "contig_abundance/all_samples_contig_abundance.txt"        --- circular plasmid abundance
    plasmid_classify/plasmid_classify.txt"                     --- circular plasmid annotation file

5. circular_non_redundant_gene:

    "circular_non_redundant_gene_rep_seq.fasta"                              --- circular plasmid gene catalog, 
    "abundance/all_circular_gene_abundance_based_contig_abundance.txt"       --- circular plasmid gene abundannce
    "functional_annotation/*"                                                --- functional annotation results of circular plasmid gene
    "annotation_ARGs/*"                                                      --- antibiotic resistance gene annotation result of circular plasmid gene
    "annotation_BacMet2/*"                                                   --- biocide- and metal-resistance genes annotation result of circular plasmid gene 
    "annotation_vf/*"                                                        --- Virulence factors genes annotation result of circular plasmid gene 

6. linear_plasmid_genome:

    "*_predict_plasmid_plasforest.fa"               ----  plasmid Contigs identified by plasforest
    "*_probs.out"                                   ----  plasmid Contigs identified by platon   

7. linear_non_redundant_contig:

    "linear_non_redundant_contigs_rep_seq.fasta"               --- linear plasmid catalog, 
    "contig_abundance/all_samples_contig_abundance.txt"        --- linear plasmid abundance

8. linear_non_redundant_gene:

    "linear_non_redundant_gene_rep_seq.fasta"                                --- linear plasmid gene catalog, 
    "abundance/all_linear_gene_abundance_based_contig_abundance.txt"         --- linear plasmid gene abundannce
    "functional_annotation/*"                                                --- functional annotation results of linear plasmid gene
    "annotation_ARGs/*"                                                      --- antibiotic resistance gene annotation result of linear plasmid gene
    "annotation_BacMet2/*"                                                   --- biocide- and metal-resistance genes annotation result of linear plasmid gene 
    "annotation_vf/*"                                                        --- Virulence factors genes annotation result of linear plasmid gene 

9. remindering_report:

    "raw_reads_to_qc_reads.txt"                  ---- reads using rate of each sample in quality control step   
    "qc_to_assembly.txt"                         ---- reads using rate of each sample in assembly step  
    "qc_to_circualr_metaplasmidspades.txt"       ---- reads using rate of each sample in assmebly circular plasmid step  
    "liner_contig_to_plasmid.txt"                ---- reads using rate of each sample in identified linear from assembly contig step
    "circualr_to_verify.txt"                     ---- reads using rate of each sample in removing non-plasmid circular genome step

10. log: the logs files of each step in plaspline.

11. temp: temp files.

Contact

author:

Wanli HE (wanli.he@bio.ku.dk)

cooperator:

Franziska Klincke (franziska.klincke@bio.ku.dk)
Joseph Nesme (joseph.nesme@bio.ku.dk)
Søren Johannes Sørensen (sjs@bio.ku.dk)