Hei @mashehu, I am working on the module tests (https://github.com/nf-core/cageseq/commit/2af5286886a408caa1cb364d96cf03e6357e0e9d) and a bit unsure what is going on with the environment file. As I understand the test automatically fetches the container, and I thought that updates automatically when the environment file is updated, but probably not because gplots is still missing ( Error in library(gplots) : there is no package called 'gplots' when tag_qc test runs) So my question is: how can I update the environment / container?

Hi @mashehu and @Kata Ferenc! Two questions: 1. I tried

time nf-test test tests/default.nf.test --updateSnapshot --profile test

on a new branch (

full-tests

) based on

dev

, using Nextflow v26.04.0 build 12031, and I'm getting loads of "undefined" and "not declared" errors about variables across the pipeline, for example:

Error subworkflows/local/mapped_inputs/main.nf:19:5: `files_in` was assigned but not declared
  │  19 |     files_in = row.path
  ╰     |     ^^^^^^^^

Error subworkflows/local/mapped_inputs/main.nf:21:9: `files` is not defined
  │  21 |     if (files.size() == 2){
  ╰     |         ^^^^^

Error workflows/cageseq.nf:18:1: Invalid include source: '/mnt/biggley/home/slava/projects/leancage_dev/cageseq/subworkflows/local/parameter_checks/main.nf'
  │  18 | include { PARAMETER_CHECKS } from '../subworkflows/local/parameter_che
  ╰     | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

I guess, this is because of the new, strict, Nextflow syntax? What are we doing with this? I can post the full list of these errors if needed. 2. Meanwhile, I'll create and populate the

<testname>.nf.test

files. In each of these files, shall I put the full list of parameters that we usually use in

.yaml

files, or just the ones that differ from the default in

cageseq/testdata/tests/params_yeast_nfcore_test_star_noindex_wfasta_se_full.yaml

? I guess, that's decided by how parameters are set by default in the schema and the pipeline itself, but I'm not sure which way we design these settings now?

One more question: in

default.nf.test

,

outdir

is set to

"$outputDir"

. Shall I retain this setting in all test files, or shall I give a unique name to

outdir

in each of them?

sorry I'll be late again

Hi @Kata Ferenc! Attached please find the whole CAGEr results directory from a full pipeline run on our nf-core test data with standard parameters. Sorry for the delay!

Hi @Kata Ferenc and @mashehu! I'm trying to run a default full-pipeline test, and it fails with the following error:

Assertion failed:

  1 of 2 assertions failed

  Nextflow stdout:

  ERROR ~ No such file or directory: Can't find a matching module file for include: ../subworkflows/local/parameter_checks/main.nf

-- Check script '/mnt/biggley/home/slava/projects/leancage_dev/cageseq/workflows/cageseq.nf' at line: 18 or see '/mnt/biggley/home/slava/projects/leancage_dev/cageseq/.nf-test/tests/645b306bac398833284fed02d9c23737/meta/nextflow.log' file for more details

subworkflows/local/parameter_checks/main.nf

does not indeed exist anymore. Do you know what I need to include in

/mnt/biggley/home/slava/projects/leancage_dev/cageseq/workflows/cageseq.nf

instead?

I'm running tests in a new

full-tests

branch which currently corresponds to

dev

(I merged

dev

into

full-tests

just now).

One more question (which we might be able to solve with @mashehu tomorrow?) -- I created a stub for a bigwig list for the

cageronly

test run:

id,single_end,path,new_name
S1,true,[?/testdata/v2/bigwig/S1.Signal.Unique.str1.out.wig.bw ?/testdata/v2/bigwig/S1.Signal.Unique.str2.out.wig.bw],S1
S2,true,[?/testdata/v2/bigwig/S2.Signal.Unique.str1.out.wig.bw ?/testdata/v2/bigwig/S2.Signal.Unique.str2.out.wig.bw],S2

However, I don't know what to put here instead of

?

which should be the location of the test data (our

test-datasets

repository).

Hi Matthias! I opened the following pull requests, as we discussed: 1. full-tests into dev: https://github.com/nf-core/cageseq/pull/110 2. Additional test data into test-datasets (for cageronly): https://github.com/nf-core/test-datasets/pull/2059

Hey @mashehu! Could you please merge my PR into dev? -- https://github.com/nf-core/cageseq/pull/110

Then I'll be able to pull from

dev

into

full-tests

again and resume testing.

Regarding the

test-datasets

, I think, the path is right -- I've put the list of samples (bigWigs/BAMs) (which is for the

cageronly

run) in a separate directory, so that it is not confused with the samplesheet (which is for the full run or

maponly

).

Hei, @Slava Sidorov mentioned that dev still doesn't work due to mandatory genome input, so I also tried and it was good for me. Did you pull the latest changes? However, what was not good is how multiqc is called. It is now expecting a single tuple with 6 entries instead of 6 channels. I checked rnaseq and there is a lot of things going on with mutliqc that we don't do yet. @mashehu do you know about how to address these changes?

on a similar note, I have a feeling that line 15 here (https://github.com/nf-core/cageseq/blame/dev/subworkflows/local/prepare_mapping_metadata/main.nf) does not do what it is supposed to, resulting in the pipeline not running through. But I did not properly debug

Hi @Kata Ferenc! Thank you for checking! I've pulled dev now and tried again, but I'm getting the same error. Could you please attach your YAML? Please find mine attached.

The error I'm getting is this:

WARN: The following invalid input values have been detected:

* --markdown_path: /mnt/biggley/home/slava/projects/leancage_dev/cageseq/assets/cager_report.Rmd


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  Genome 'reproduce_test_data/sacCer3chrIII.fa' not found in any config files provided to the pipeline.
  Currently, the available genome keys are:
  GRCh37, GRCh38, CHM13, GRCm38, TAIR10, EB2, UMD3.1, WBcel235, CanFam3.1, GRCz10, BDGP6, EquCab2, EB1, Galgal4, Gm01, Mmul_1, IRGSP-1.0, CHIMP2.1.4, Rnor_5.0, Rnor_6.0, R64-1-1, EF2, Sbi1, Sscrofa10.2, AGPv3, hg38, hg19, mm10, bosTau8, ce10, canFam3, danRer10, dm6, equCab2, galGal4, panTro4, rn6, sacCer3, susScr3
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

genome

must be a valid igenomes name, (see the list given in the error). in you case you should be able to leave it blank and just provide

fasta

Thank you Matthias! Leaving

genome

empty and setting

fasta

, I'm now getting this one:

WARN: The following invalid input values have been detected:

* --markdown_path: /mnt/biggley/home/slava/projects/leancage_dev/cageseq/assets/cager_report.Rmd


WARN: Found the following unidentified headers in /mnt/biggley/home/slava/projects/leancage_dev/test_samplesheet/samplesheet.csv:
        - single_end

Reading in samplesheet
Creating channel from samplesheet
Initializing channels
ERROR ~ Reference genome FASTA file (--genome) or genome index (--index) should be specified.

 -- Check '.nextflow.log' file for details

huh, that's weird, can you attach your yaml?

i actually didn't use a yaml just command line defining some parameters including --fasta

Hi @Kata Ferenc, sure, here it is.

Just to make sure we are using the same version of the pipeline -- I'm in

dev

, using the following commit:

$ git log | head -15
commit b40cc0408eba183363b2e59fa86176912b358018
Merge: f3ce065 50ea4d8
Author: Matthias Hörtenhuber <mashehu@users.noreply.github.com>
Date:   Tue May 26 17:07:42 2026 +0200

    Merge pull request #106 from nf-core/nf-core-template-merge-4.0.2

    Important! Template update for nf-core/tools v4.0.2

(I've pulled 5 min ago.)

Hi all, I started a bit of debugging and one thing that came up is that the new module upgrades are not bacward compatible with nextflow 25.10.4. I think we need to reflect it in the requirements. What do you think @mashehu?

looks like there was still some remnants from the

genome

==

fasta

confusion in the

test_full

PR. tried to fix that now.

Hi @mashehu! Thank you! I pulled-rebased

test_full

(apparently, I had some unpushed commits on my side), and now it gives the same error as the one described by Kata in the thread above:

-[nf-core/cageseq] Pipeline completed with errors-
WARN: Singularity cache directory has not been defined -- Remote image will be stored in the path: /mnt/scratch/slava/cageseq_test/singularity -- Use the environment variable NXF_SINGULARITY_CACHEDIR to specify a different location
WARN: Input tuple does not match tuple declaration in process `NFCORE_CAGESEQ:CAGESEQ:BAM_STATS_SAMTOOLS:SAMTOOLS_STATS` -- offending value: [sacCer3chrIII, /mnt/biggley/home/slava/projects/leancage_dev/reproduce_test_data/sacCer3chrIII.fa]
ERROR ~ No such variable: task

 -- Check script 'cageseq/modules/nf-core/samtools/stats/main.nf' at line: 6 or see '.nextflow.log' file for more details
ERROR ~ Pipeline failed. Please refer to troubleshooting docs for common issues: <https://nf-co.re/docs/running/troubleshooting>

 -- Check '.nextflow.log' file for details

(Plus a warning about input and output of the

samtools stats

module.)

And, just in case, the current

dev

breaks with this:

Reading in samplesheet
Creating channel from samplesheet
Initializing channels
ERROR ~ Reference genome FASTA file (--genome) or genome index (--index) should be specified.

 -- Check '.nextflow.log' file for details

The YAML is as follows:

# pipeline parameters
fullpipeline: true
maponly: false
cageronly: false
gtf: "reproduce_test_data/sacCer3chrIII.gtf"

# preprocessing parameters
input: "/mnt/biggley/home/slava/projects/leancage_dev/test_samplesheet/samplesheet.csv"
infolder:
outdir: "results_nfcore_standard_samplesheet"
sample_name_fields: 1
genome_name: "sacCer3chrIII"
genome:
fasta: "reproduce_test_data/sacCer3chrIII.fa"
index:
seq_platform: "illumina"
seq_center: false
# whether to take the uniquely mapped only
unique_only: true
# whether to remove reads that do not start with G
remove_non_g: false
# whether to trim G or not
nogtrim: false
# mapper to use
bowtie2: false
# dedup
dedup: false

# CAGEr parameters
cager_sample_file: ../cageseq2_sample_list_test.csv
# datatype: "bam" or "bigwig". if bowtie2 is used, it will be set to bam
datatype: "bigwig"
# BSgenome
forgeseed:
sourcedir:
bsgenome: "reproduce_test_data/BSgenome.sacCer3ChrIII.Test.sacCer3ChrIII_1.0.tar.gz"
# parameter for correlation calculation
corrplot_tagCountThreshold: 1
# parameters for normalization
norm_range_min: 5
norm_range_max: 10000
norm_method: "powerLaw"
alpha:
t_norm: 1000000
# parameters for tag clustering
sample_num_thr: 2 # 1
ctss_thr: 3 # 1
distclu_maxDist: 20
keepSingletonsAbove: 1
iq_low: 0.1
iq_high: 0.9
# plotting for tagclusters QC
iqw_tpm_threshold: 3
tssregion_up: -3000
tssregion_down: 3000
tsslogo_upstream: 35
# parameters for consensus clusters
consensus_thr: 2
consensus_dist: 20
# parameters for enhancer calling
cfBalanceThreshold: 0.65 # 0.95
unexpressed: 0
minSamples: 0

Hi @mashehu and @Kata Ferenc! Please let me know if I can help with anything! (I which case, I would need detailed instructions 🙂).