I have a question about DMR calling with three cell lines (groups), each with 4–5 replicates. In the documentation, I saw the options

--dmr_population_scale

and

--population_dmrer

, but I’m unsure whether these are meant for much larger “population-scale” experiments and if they are appropriate for my setup. I’d also like to compare DMRs in one cell line against all others. Is there a way to do this within a single run, or do I need to specify

--dmr_a

and

--dmr_b

for each pair (e.g. three separate runs to cover all combinations)? In other words, can the pipeline perform all pairwise comparisons between 3 groups in one go, or must they be run separately? And what options would I have to set for that in my script? Any guidance on recommended settings for this design would be greatly appreciated.

Hi @Magdalena Winklhofer, thank you for your question!

--dmr_population_scale

is used to indicate whether to perform DMR analysis for population-scale.

--population_dmrer

is for specifying which DMR tool to use for population-scale analysis, default is dss. Unfortunately, methylong currently only supports DMR analysis between two groups at a time. If you have three groups, you would need to run the analysis three times, specifying different pairs for each comparison. We understand this might not be the most convenient for your design, and we will consider further improvements to support multi-group comparisons in future versions of methylong.

Hi @Yilia X, thank you for your response. I now ran the pipeline with the following settings: nextflow run nf-core/methylong/ \ -revision 2.0.0 \ -profile singularity \ --input "${WORKDIR}/samplesheet/samplesheet_cell_lines.csv" \ --outdir "${WORKDIR}/results" \ --no_trim \ --dmr_population_scale \ --dmr_a cell_line1 \ --dmr_b cell_line2 \ -work-dir "${WORKDIR}/work" \ -resume \ -with-report "${WORKDIR}/logs/nextflow_report.html" \ -with-trace "${WORKDIR}/logs/nextflow_trace.txt" \ -with-timeline "${WORKDIR}/logs/nextflow_timeline.html" The pipeline finished successfully. And I would like to see the differences in methylation between the cell lines that I tested. Hence, I would like to identify the DMR between cell line 1 and cell line 2. You listed in your documentation that the output of the DMR analysis should contain: modkit dmr output: •

dmr_haplotype_level/modkit/

• `*_2|etc.bed`: partitioned reads based on HP tag • `*_modkit_dmr_haplotype_level.bed`: differential methylation output •

dmr_population_scale/

• `*_DSS_DMLtest.txt`: DML test results • `*_DSS_callDML.txt`: DML • `*_DSS_callDMR.txt`: DMR • `*_DSS.log`: DSS log But after completing the pipeline, I only found results/ont/cellLine1_cellLine2_directory/dmr_population_scale/dss: which contains a log file. Additionally, I found under ont/cell_line1_sample_1/dmr_population_scale/modkit/ a bed file, but no `*_DSS_callDMR.txt`: DMR as described in the documentation. Could you let me know if I made a mistake while running the pipeline, set a wrong option or looked in the wrong place to find my results? Additionally, I tried to download the .bed files and load them into igv viewer, but got a error, that "Unexpected data in columns 10-12, expected exon count, exon starts, and exon sizes. Found: 1 100.00 1 Columns > 9 will be ignored". Is that normal? DO I need to perform a conversion for the bed files to be loadable into igv? Thank you very much for your help!

@Magdalena Winklhofer Could I take a look at your log file in

results/ont/cellLine1_cellLine2_directory/dmr_population_scale/dss

? I think something went wrong with the

DSS

.

@Yilia X: Yes, you could. Could I send you those on a private email, since the data that I am working on should not be viewed publicly?

@Magdalena Winklhofer Sure! You can send them to yshico@163.com.

@Yilia X, could you let me know if you received my email?

@Magdalena Winklhofer Sorry for replying late. I’ve emailed you a reply, please check it at your convenience.

Hi, I am having similar issues in which only a DSS .log file is created but no output. Any ideas?