Diffbind bioconductor, peaks overlapping

Question

[marina-orlova]

Hello

Can you please help me with diffbind package.

As I understood I should create a csv file with names of bed files with input, control and peaks bed files of peaksets.

Are all of these files required for analysis? Can I do only with files with peaks? I'm trying to do without initial files (just don't have them) writing a csv file as follows:

SampleID,Peaks
1,2A.peaks.bed
2,3A.peaks.bed
3,5A.peaks.bed

But using this in R I have a mistake:

> sets = dba(sampleSheet="sample.csv")
1     NA raw
Error in Summary.factor(c(1L, 1112L, 2223L, 3334L, 4445L, 5556L, 6592L,  : 
  max not meaningful for factors
Warning:
In read.table(samplesheet, sep = ",", stringsAsFactors = F, header = T) :
  incomplete final line found by readTableHeader on 'sample.csv'

[nirmalyasaha2003]

Hi Marina,

I was facing similar problem. I think, unknowingly you got some spaces while writing the CSV file. Open the file in a texteditor (e.g Textedit for MAC or Notepad for Windows) and then check if there are any unwanted spaces.

The file should somewhat look like this

SampleID,XXX,YYY,AAA

Not like this

SampleID,XXX, YYY,AAA

Even check the end of a line to see any extra spaces, which is visually unrecognizable

Thanks

[marina-orlova]

Hi, thank you for the reply, I managed to make it work )

[zwang0614]

Hi Marina-orlova, May you please tell how you made it work? I met the same problem now. Thanks.

[Jens Lichtenberg]

Hi zwang,

You need to make sure that your peak file is formatted correctly. It must be in BED format and follow this form:

CHROMOSOME   START   END   SCORE