Problems with long repetitive regions

[miespejodiaz]

Hi Jeremy Marcus,

First off, thank you for providing the tool, it's incredibly useful!

I’ve encountered a small issue related to the representation of highly repetitive sequences in my reference genome. Specifically, my reference includes the first and last 50K base pairs of each chromosome, with around 20K repetitions of the telomere region ("TTAGGG" or "AATCCC"). While the tool works great overall, when I try to represent a region that contains this repetitive sequence (20K repeats) plus a consecutive 20K base pair non-repetitive intrachromosome region, the tool seems to cut off the repetitive sequence representation at some point.

I would like to include the full repetitive region in the plots, as it would be valuable for my analysis to shown each read length together with the methylation (similar to the visualization at IGV). I'm specifically interested in shown how the methylation data could varies based on telomere read length for each read.

Could you help me find a way to ensure that the entire region, including the repetitive sequences, is included in the plots?

Thanks again for your great work on this tool!

Best regards

Isa

[OberonDixon]

Hi Isa,

I'm sorry this question slipped through the cracks! I am not immediately sure what is going wrong in the plots above; I do see that the length of plotted read is shorter in dimelo than in IGV. What I can tell you is that our software simply goes based on the alignment in the .bam file; if things are missing from the plot_reads output I believe that must be indicative of the information in the underlying .bam file. If you look at plot_read_browser you could examine the read extents (start and end of alignment) and that might inform whether past some point into the telomere there just aren't modified bases in the reads.

[OberonDixon]

Hi Isa,

We just released the new v0.2.1 which has some small adjustments to handling for reads with large alignment gaps in them. I am not certain whether this will interact with your issue, but it could be worth trying to see!