Hello all,
I'm relatively new to bioinformatics and am currently analyzing CUT&RUN-seq of histone markers with Diffbind. Some of the questions or things I want to try to answer with Diffbind are how my samples cluster and what exact affinity sites are specific to each group. I'm able to run Diffbind on my samples with the following code:
mydba <- read.csv("samples.csv")
mydba <- dba(sampleSheet = mydba , peakCaller = "macs", peakFormat = "macs", scoreCol = 7 )
mydba <- dba.blacklist(mydba, blacklist= DBA_BLACKLIST_MM10)
mydba <- dba.count(mydba)
mydba <- dba.normalize(mydba)
mydba <- dba.contrast(mydba, minMember=2,
reorderMeta = list(Condition= "Naive"))
mydba <- dba.analyze(mydba)
Afterwards, I am able to generate heatmaps showing the clustering of the samples with the following code:
hmap <- colorRampPalette(c("blue", "white", "red"))(n = 13)
readscores <- dba.plotHeatmap(mydba, correlations=FALSE,
scale="row", colScheme = hmap)
This produces the following image:
The heatmap is clustering as I expect, but I wanted to know more in depth about the regions used on the y-axis. I was wondering with the Diffbind pipeline, or any method, if there is a way to identify and/or export specific regions on the y-axis to help answer my question above. Additionally, I was wondering if there are resources for choosing acceptable color palletes for data visualization (e.g., for color blindness). Thank you all for your help in advance.
