metagenomicReport

A workflow for checking Fastq files for possible contamination with reads from species other than human, primarily cell culture samples

Overview

Dependencies

• kraken2 2.0.8
• bracken 2.7

Usage

Cromwell

java -jar cromwell.jar run metagenomicReport.wdl --inputs inputs.json

Inputs

Required workflow parameters:

Parameter	Value	Description
fastqR1	File	Fastq R1
fastqR2	File	Fastq R2
outputPrefix	String	Output is, usually sample name

Optional task parameters:

Parameter	Value	Default	Description
krakenReport.modules	String	"kraken2/2.0.8 kraken2-pluspf-database/1"	Names and versions of modules needed for read classification
krakenReport.krakenDb	String	"$KRAKEN2_PLUSPF_DATABASE_ROOT/"	Path to bracken/kraken db
krakenReport.krakenOut	String	"/dev/null"	Redirect kraken2 output, default is /dev/null
krakenReport.timeout	Int	24	Timeout in hours for this task
krakenReport.jobMemory	Int	20	Java memory for Kraken
brackenReport.modules	String	"bracken/2.7 kraken2-pluspf-database/1"	Names and versions of modules needed for read ratio estimation
brackenReport.krakenDb	String	"$KRAKEN2_PLUSPF_DATABASE_ROOT/"	Path to bracken/kraken db
brackenReport.classLevel	String	"S"	Classification level, default S (species)
brackenReport.readLength	Int	100	Expected read length
brackenReport.threshold	Int	10	minimum number of reads required for a classification
brackenReport.timeout	Int	24	Timeout in hours for this task
brackenReport.jobMemory	Int	20	Java memory for Bracken
brackenReport.minRatio	Float	0.03	Threshold for reporting species, minimum read proportion in the analyzed sample

Outputs

Output	Type	Description	Labels
textReport	File	a report text file generated by Bracken	vidarr_label: textReport
jsonReport	File	json report with bracken-collected estimates	vidarr_label: jsonReport

Commands

This section lists command(s) run by metagenomicReport workflow

• Running metagenomicReport

Kraken2

Kraken2 accepts paired fatq files and runs classificatiion of reads using k-mer database of various bacterial, fungal, protozoan and viral species.

kraken2 --paired FASTQ_R1 FASTQ_R2 
        --db KRAKEN_DB 
        --report SAMPLE.kreport2.txt 
        --output /dev/null

Bracken

Bracken (Bayesian Reestimation of Abundance with KrakEN) is a highly accurate statistical method that computes the abundance of species in DNA sequences from a metagenomics sample. Braken uses the taxonomy labels assigned by Kraken, a highly accurate metagenomics classification algorithm, to estimate the number of reads originating from each species present in a sample.

bracken -d KRAKEN_DB 
        -i KRAKEN_REPORT 
        -o SAMPLE.bracken 
        -r READ_LENGTH 
        -l CLASS_LEVEL 
        -t THRESHOLD

python3<<CODE
import json
import json

report_file = ~{sample}.bracken
json_name = ~{sample}_brackenReport.json
sampleName = ~{sample}
jsonDict = {sampleName: []}
header = []
limit = ~{minRatio}    # minimum read fraction to consider as contamination

"""For Bracken, we need fields 2,4,5,6 to be int and 7 - float type"""
def typeCast(reportString):
stringList = reportString.split("\t")
if len(stringList) != 7:
    return stringList
for i in [1, 3, 4, 5]:
    stringList[i] = int(stringList[i])
stringList[6] = float(stringList[6])
return stringList

"""Read from Bracken report, convert to json"""
with open(report_file) as r:
for line in r:
    lineIn = line.rstrip()
    if lineIn.find("taxonomy_id") > 0:
	header = lineIn.split("\t")
	continue
    tmp = typeCast(lineIn)
    if float(tmp[-1]) < limit:
	continue
    jsonDict[sampleName].append(dict(zip(header, tmp)))
r.close()

if len(jsonDict.keys()) > 0:
with open(json_name, 'w') as json_file:
    json.dump(jsonDict, json_file)
CODE

Support

For support, please file an issue on the Github project or send an email to [email protected] .

Generated with generate-markdown-readme (https://github.com/oicr-gsi/gsi-wdl-tools/)