I am very new to ChipSeq and am just so confused. I have my sequence data, the control and the reference from NiH and UCSC. I have successfully (I think) , mapped the reference, aligned them with bowtie and found peaks with MACS2. The output of MACS2 was a .r file, a .xls file, a .narrowPeak file and a .bed file. I now want to move onto finding differentials. Everything I read watch tells me I need a sample sheet. Is this the .xls file? Or do I need to create something else? Right now this is my R script:

library(ChIPQC)
library(DiffBind)
library(tidyverse)

sampleSheet <- read.csv("meta/peaks.xls")
db <- dba(sampleSheet =sampleSheet)

When I look at the created object the output is:

15741 Samples, 0 sites in matrix

Obviously I'm doing something wrong. Any help would be greatly appreciated.

The sample sheet is a separate .csv file that contains a row for each sample, with columns indicating the locations of the peak and bam files, as well as experimental metadata. The columns are document in the help page for ?dba.

You should think about working through the example in the vignette before you attempt your own analysis. You can see the example sample sheet as follows:

samplesheet <-  system.file("extra/tamoxifen.csv", package="DiffBind")
samples <- read.csv(samplesheet)
samples

For the peak files, you can use either the .xls files (PeakCaller="macs") or the .narrowPeak files (PeakCaller="narrow").