I am analyzing ChIP-seq data for a common protein target across three distinct cell lines. My objective is to perform a comparative analysis of peak profiles. Currently, I understand that if I want to use the

nf-core/chipseq

pipeline for this task, I would need to generate individual

samplesheet.csv

files for each cell line's experiment and execute three separate runs. So, I could compare the resulting

.bed

files. However, I wonder if I can consolidate the

.fastq.gz

files from all three experiments into a single

samplesheet.csv

and conduct a single pipeline run instead of three? Thank you!

Hi Dimitri, you can definitly run all you cell lines at once, you just have to build your samplesheet accordingly (chipseq usage). You also have example of such samplesheet for paired-end and single-end reads here : chipseq github.

Thank you. Would that mean I will have three different .bed files (for each experiment) in the end? Thank you again

Yes ! I think all you need is to specify that you samples are not replicates.

I am sorry to ask about it again, but is there currently a way to use the T2T-CHM13 genome assembly as a predefined option in the nf-core/chipseq pipeline? I have provided them as a separate .fna and .gtf, that were downloaded before, but is there a --genome option by chance? Thank you

So at the moment

--genome

refer to igenome listed here : https://ewels.github.io/AWS-iGenomes/. I don't think yours is in the list but you can check. Honnestly I never been able to use this

--genome

parameter, so I just download the igenome files and give it as separated parameters such as :

--fasta references/Drosophila_melanogaster/UCSC/dm6/Sequence/Bowtie2Index/genome.fa \
 --gtf references/Drosophila_melanogaster/UCSC/dm6/Annotation/Genes/genes.gtf \
  --bowtie2_index references/Drosophila_melanogaster/UCSC/dm6/Sequence/Bowtie2Index/

It should work the same for external genomes. So what you would need is to index your reference genome with bowtie2 and give the path such as the exemple above.