Nextflow Conversion of ASVTableTask.pm

MarkerGeneAnalysis16sDADA2

flowchart TD
    p0((Channel.fromList))
    p1[markerGeneAnalysis:downloadFiles]
    p2[markerGeneAnalysis:filterFastqs]
    p3[markerGeneAnalysis:buildErrors]
    p4[markerGeneAnalysis:fastqToAsv]
    p5[markerGeneAnalysis:mergeAsvsAndAssignToOtus]
    p6(( ))
    p7(( ))
    p8(( ))
    p0 -->|ids| p1
    p1 --> p2
    p2 --> p3
    p3 --> p4
    p4 --> p5
    p5 --> p8
    p5 --> p7
    p5 --> p6

Loading

Decription of nextflow configuration parameters:

param	value type	description
studyIdFile	path	Path to SRAID tsv file
platform	string	Platform the sequencing was done on. Used in filterFastqs step. Ex: "illumina"
isPaired	boolean	Is the data paired?
trimLeft	int	Amount to trim from left. Used in filterFastqs step.
trimLeftR	int	Amount to trim from reverse strand. Used in filterFastqs step.
truncLen	int	Amount to truncate from forward strand. Used in filterFastqs step.
truncLenR	int	Amount to truncate from reverse stand. Used in filterFastqs step.
maxLen	int	Maximum length. Used in filterFastqs step.
mergeTechReps	boolean	Would you like to merge TechReps in fastqToAsv step.
trainingSet	path	Path to training set fasta.
speciesAssignment	path	Path to species assignment fasta.
outputDir	path	Path where you would like results to be stored
nValue	scientific notation	Ex: 1e+02 Used in buildErrors Step. Setting value above 1e+02 can cause issues with memory

In order to run the script, you will need to supply to the nextflow.config file:

1. An apikey to be used to collect the fastq files from NCBI
2. The location of your training set file (a short example file has been supplied in /data)
3. The location of your species assignment file (a short example file has been supplied in /data)

The parameter names for these have already been specified, you only need to enter them into the empty quotes.

The output files from this tool will be formatted as (SRRID supplied in sraStudyIDFile parameter)_(outputName parameter).(nothing, "bootstraps", or "full")

Get Started

• Install Nextflow

  curl https://get.nextflow.io | bash

• Run the script

  nextflow run VEuPathDB/MarkerGeneAnalysis16sDADA2 -with-trace -c <config_file> -r main

Running this locally can commonly result in an OOME. This results from the derepFastq function in buildErrorsN.R having n as too great a value.

drpsF[[i]] <- derepFastq(inputFilesF[[i]], n = 1e+06, verbose = verbose)

The default n is 1 million reads. If an OOME occurs, a lower number can be entered.

Ex: drpsF[[i]] <- derepFastq(inputFilesF[[i]],n = 1e+03, verbose = verbose)

This is specified in the nextflow.config file parameter "nValue".

Documentation for derepFastq and other dada2 functionality can be found here.