Program to View Average Coverage with Variance of BAM/Bigwig Files?

Question

[vanbelj]

I am viewing BAM/bigwig files from ChIP-seq experiments. Every experiment has 3 biological replicates. Are there any tools that will plot the average coverage from replicates with a measure of variance? I have too many conditions and viewing any more than 9 individual tracks just isn't feasible.

I'm sure I could create a 'dummy' BAM/bigwig where I calculate the average coverage myself, and that would be compatible with viewers such as IGV. However, I'd really like to have a measure of variance included in the plot so I am alerted of potential issues with reproducibility amongst replicates.

Are there any programs that can do this? Thanks!

Answer 1

[cmdcolin]

ChIP-seq is difficult because it inherently has lots of variance (noise-y+peak-y signals). This program is a cool example but is focused on CNV https://brwnj.github.io/covviz/#loc=158307649-168161122

I also catalog a lot of tools which you can browse to maybe get inspiration, but nothing i know of exactly fits your bill https://cmdcolin.github.io/awesome-genome-visualization/

You may need to manually define what you mean by variance (e.g. per-base variance, or variance within a sliding window, or just highlighting presence/absence of a peak call, etc) and this may help guide yourself to a solution that you can maybe even roll yourself :)

[vanbelj]

I was thinking per-base variance, though sliding window would also work to alert of high variance between replicate. Using R, I make average coverage plots ± variance for individual genes, but I do not have a tool/GUI to actively browse the genome. So, it requires me to identify interesting genes through other mechanisms and then create plots of said genes. It's pretty wasteful: time consuming to run, occupies lots of HD space, and end up pitching many plots that end up being uninteresting.

Thank you for the suggestions about tools. I have no experience in GUI development and don't think that's a journey I'll embark on at the moment.