plotBrowserTrack() results look truncated

[graciegordon]

Hello @rcorces and @jgranja24,

I have a follow up question to discussion #903. Is there a way to specify ylims with the signal range rather than the quartile? Using the quartiles with ylims=c(0,1) looks like it might be truncating some of the peaks (ex: NK-B-2). I am using the archR v1.

.

[rcorces]

I dont have enough information to help you. ArchR "v1" is not specific. As you indicated #903 discusses a bug in the ylim argument. This fix has been implemented in the release_1.0.2 branch which you can install via:
devtools::install_github("GreenleafLab/ArchR", ref="master", repos = BiocManager::repositories())
Then restart your R session and reload ArchR with library(ArchR)

Make sure you are using the correct parameter (ylim not "ylims") and note that if ylim is not set, ArchR defaults to using the 0.999 quantile instead of 1.

ArchR/R/ArchRBrowser.R

Line 997 in ef6fd03

if(!is.null(ylim)){

It is possible that your "truncating" of the axis is due to:

1. using a version of ArchR that doesnt have the above-mentioned fix implemented
2. incorrect usage of the ylim param
3. Smoothing of the signal that makes it look like the y value is higher than the y axis limit.