Hi everyone,
I’m trying to run nf-core/sopa on my MACSima imaging samples.I’m working on a RedHat HPCC environment, using:
Nextflow version: 25.10.0
nf-core/sopa revision: master (commit 11b4cccb7e)
Container engine: Singularity
My config.yml
# For parameters details, see this commented example: https://github.com/gustaveroussy/sopa/blob/main/workflow/config/example_commented.yaml
read:
technology: macsima
patchify:
patch_width_pixel: 3000
patch_overlap_pixel: 40
segmentation:
cellpose:
diameter: 35
channels: [ "DAPI" ]
flow_threshold: 2
cellprob_threshold: -6
min_area: 400
aggregate:
aggregate_channels: true
min_intensity_ratio: 0.1
expand_radius_ratio: 0.1
explorer:
ram_threshold_gb: 4
pixel_size: 0.170
My samplesheet.csv
sample,data_pathROI1,/storage/scratch01/groups/lc/rnafusion_pdx/sopa/sample1/ROI2,/storage/scratch01/groups/lc/rnafusion_pdx/sopa/sample2/
The directory contains the MACSima output (multiple TIFFs).
Command used
/home/jgracia/bin/nextflow run nf-core/sopa -profile singularity --input samplesheet.csv -params-file config.yml --outdir /storage/scratch01/groups/lc/rnafusion_pdx/sopa/ --email josearacena111@gmail.com
The problem
ERROR ~ Validation of pipeline parameters failed!Entry 1: Missing required field(s): fastq_1Entry 2: Missing required field(s): fastq_1
Even though I explicitly set:
read: technology: macsima
This doesn’t make much sense since MACSima doesn’t produce reads of any kind, and since the tecnology used is specified I’m not sure how to solve the error.
"This is the development version of the pipeline". You should ask the dev on slack "For further information or help, don’t hesitate to get in touch on the Slack #sopa channel (you can join with this invite)."
Thank you! But there's something I might be missing because I cant seem to find any invitation to the sopa slack channel
To join slack, you can use the link for slack on nf-core. And then join the #sopa channel, either directly on slack or with this link
The spambot got triggered here, sorry. Restored your post. I will reach out to the admin to see whether we can train it to not confuse the nf-core links and channel hashes with spam.
yeah, no worries, I think OP got there in the end