Closed roskamsh closed 3 weeks ago
nf-core pipelines lint
overall result: Passed :white_check_mark: :warning:Posted for pipeline commit a885116
+| ✅ 317 tests passed |+
#| ❔ 7 tests were ignored |#
!| ❗ 4 tests had warnings |!
Hey @roskamsh, thanks for taking over this issue! Nice work, just a couple of small changes.
@suzannejin I just had a quick look at the output and was wondering if you know why the DESeq2 result tables differ slightly in the test_experimental
profile when compared to the test
profile:
test:
gene_id baseMean log2FoldChange lfcSE pvalue padj
ENSMUSG00000023978 309.50335903 5.015358 2.7289078 NA NA
ENSMUSG00000030324 1881.77922741 0.9717132 2.0668858 NA NA
ENSMUSG00000031450 195.24652216 0.8356462 1.90527993 NA NA
experimental:
gene_id baseMean log2FoldChange lfcSE pvalue padj
ENSMUSG00000023978 308.89789121 5.406014 2.57197697 NA NA
ENSMUSG00000030324 1763.85049182 1.057522 2.14925341 NA NA
ENSMUSG00000031450 201.02654536 1.131211 2.19318651 NA NA
Hey @roskamsh, thanks for taking over this issue! Nice work, just a couple of small changes.
@suzannejin I just had a quick look at the output and was wondering if you know why the DESeq2 result tables differ slightly in the
test_experimental
profile when compared to thetest
profile:test:
gene_id baseMean log2FoldChange lfcSE pvalue padj ENSMUSG00000023978 309.50335903 5.015358 2.7289078 NA NA ENSMUSG00000030324 1881.77922741 0.9717132 2.0668858 NA NA ENSMUSG00000031450 195.24652216 0.8356462 1.90527993 NA NA
experimental:
gene_id baseMean log2FoldChange lfcSE pvalue padj ENSMUSG00000023978 308.89789121 5.406014 2.57197697 NA NA ENSMUSG00000030324 1763.85049182 1.057522 2.14925341 NA NA ENSMUSG00000031450 201.02654536 1.131211 2.19318651 NA NA
Could this be due to the new parameter I added in the test_experimental.config file? When the DESEQ2 process ran as initially specified, this line (https://github.com/nf-core/differentialabundance/blob/dev-ratio/modules/nf-core/deseq2/differential/templates/deseq_de.R#L486) threw an error, saying that nsub was greater that nrow. I went in and replicated the issue myself, and it seems that the input matrix only contained 982 features, whereas the vst_nsub
default value was set to 1000. In order to resolve this I had to change the nextflow parameter vst_nsub
in the test_experimental.config file. I am not really sure why this wasn't found in the test
run itself, if the input data is exactly the same?
Is the input data exactly the same? Even so, this would only really change the *.vst.tsv
output and not the *.deseq2.results.tsv
output.
Hey @roskamsh, thanks for taking over this issue! Nice work, just a couple of small changes. @suzannejin I just had a quick look at the output and was wondering if you know why the DESeq2 result tables differ slightly in the
test_experimental
profile when compared to thetest
profile: test:gene_id baseMean log2FoldChange lfcSE pvalue padj ENSMUSG00000023978 309.50335903 5.015358 2.7289078 NA NA ENSMUSG00000030324 1881.77922741 0.9717132 2.0668858 NA NA ENSMUSG00000031450 195.24652216 0.8356462 1.90527993 NA NA
experimental:
gene_id baseMean log2FoldChange lfcSE pvalue padj ENSMUSG00000023978 308.89789121 5.406014 2.57197697 NA NA ENSMUSG00000030324 1763.85049182 1.057522 2.14925341 NA NA ENSMUSG00000031450 201.02654536 1.131211 2.19318651 NA NA
Could this be due to the new parameter I added in the test_experimental.config file? When the DESEQ2 process ran as initially specified, this line (https://github.com/nf-core/differentialabundance/blob/dev-ratio/modules/nf-core/deseq2/differential/templates/deseq_de.R#L486) threw an error, saying that nsub was greater that nrow. I went in and replicated the issue myself, and it seems that the input matrix only contained 982 features, whereas the
vst_nsub
default value was set to 1000. In order to resolve this I had to change the nextflow parametervst_nsub
in the test_experimental.config file. I am not really sure why this wasn't found in thetest
run itself, if the input data is exactly the same?Is the input data exactly the same? Even so, this would only really change the
*.vst.tsv
output and not the*.deseq2.results.tsv
output.
I thought so too but tested the two profiles with the same nsub values and the difference was still there. But I just realized that in the experimental config, no --transcript_length_matrix
is provided. I tested it, that is indeed the source of the differences, so all is good :)
Thanks for the work!
@roskamsh thank you so much for the contribution! And thank you @WackerO for reviewing! Lgtm too! I will merge now.
PR checklist
nf-core lint
).nf-test test main.nf.test -profile test,docker
).nextflow run . -profile debug,test,docker --outdir <OUTDIR>
).docs/usage.md
is updated.docs/output.md
is updated.CHANGELOG.md
is updated.README.md
is updated (including new tool citations and authors/contributors).