gis-rpd / rpd-rnaseq Goto Github PK

repeated arg in collectrnaseqmatrics rule fail the pipeline
${arg}
need to comment this line ##### # STRAND=SECOND_READ_TRANSCRIPTION_STRAND
Line No. 379

See #9

multiqc trimmgalore report collect missing
not sure why multiqc report only one sample of collectRnaSeqmatrics *RNA_Metrics.txt (AAHHT013)
Please see the results and multiqc report file here..

"/mnt/projects/rpd/devs/justin/rpd-rnaseq-test/results/MultiQC/multiqc_report.html"

repeated arg in collectrnaseqmatrics rule fail the pipeline
${arg}
need to comment this line ##### # STRAND=SECOND_READ_TRANSCRIPTION_STRAND
Line No. 379
3.
Linne No. 388
change rseqc results outside the sample_id folder.

Leave note on how to use publishdir and workdir (and how to cleanup) in docs

Linne No. 388
change rseqc results outside the sample_id folder.

See #9

see nf-core/rnaseq#119

No sample based folder per analysis
push all the samples published results to relevant folders

bigwig/
_fwd.bw
_rev.bw

RnaSeqMetrics/
_RNA_Metrics.txt

rsem/
.genes.results
.isoforms.results
.pdf
.stat

rseqc/
plot.pdf
dup_pos/.pos.DupRate.xls
dup_seq/seq.DupRate.xls
rscripts/plot.r
read_distribution/.read_distribution.txt
bam_stat/.bam_stat.txt

STAR/
bam/_Aligned.sortedByCoord.out.bam
bam/_Aligned.toTranscriptome.out.bam

count/_ReadsPerGene.out.tab
logs/Log.final.out
logs/SJ.out.tab
logs/Log.progress.out

move fastqc to out/ folder

move MultiQC to out/ folder

keep the

remove the following under "Library strandedness "
(Reference format https://github.com/nf-core/rnaseq/blob/master/docs/usage.md)
"""
Specifying --pico makes the pipeline run in forward_stranded mode.

You can set a default in a custom Nextflow configuration file such as one saved in ~/.nextflow/config (see the nextflow docs for more). For example:

params {
reverse_stranded = true
}
If you have a default strandedness set in your personal config file you can use --unstranded to overwrite it for a given run."
""""

Include the below..
"""
These flags affect the commands used for several steps in the pipeline - namely RSEM and picard collectRnaSeqMetrics
:
RSEM:

--unstranded none
--forward_stranded forward
--reverse_stranded reverse (Illumina TruSeq Stranded protocols)

Picard collectRnaSeqMetrics:

--unstranded STRAND_SPECIFICITY=NONE
--forward_stranded || singleend STRAND_SPECIFICITY=FIRST_READ_TRANSCRIPTION_STRAND
--reverse_stranded STRAND_SPECIFICITY=SECOND_READ_TRANSCRIPTION_STRAND

""""

not sure why multiqc report only one sample of collectRnaSeqmatrics *RNA_Metrics.txt (AAHHT013)
Please see the results and multiqc report file here..

"/mnt/projects/rpd/devs/justin/rpd-rnaseq-test/results/MultiQC/multiqc_report.html"

See #9

• Mention change of workdir and publishdir
• Make sure all options are in docs or refer to online usage
• Mention that options ideally go into yaml
• mention submission of Nextflow itself

1. By default, run fastqc on raw data and collect for Multiqc
   If trimming required, pass only trimming report for MultiQC
2. Publish for rsem ($sampleID}.stat folder
3. Enable rsem and collectRnaseqMetrics into MultiQC
4. picard strand information (optional)
5. CollectRnaSeqmetrics - add strand specificity (Refer https://github.com/gis-rpd/pipelines/blob/devel/rnaseq/star-rsem/Snakefile)

esp. remove CPU settings for programs not using them or adjust for those not using them efficiently (FastQC, MultiQC, Picard)

fastqc html and zip files were not written in the publish dir

/mnt/projects/rpd/devs/justin/rpd-rnaseq-test/results/fastqc

Logging activity needs to bemonitored