Hi 

I am new to R and to DiffBind. I am trying to analyse some ChipSeq data. I have aligned reads to the genome and obtained a bam and bed file. Then I obtained peak calls using MACS.

While using diffbinds, I am getting some error message with the dba command. 

The error message is as follows :- 

> test <- dba(sampleSheet="samples.csv")

K1 white NT Media Treated 1 raw

Error in Summary.factor(c(1L, 11112L, 22223L, 28200L, 29311L, 30422L,  :

  ‘max’ not meaningful for factors

The first four lines of my samples.csv file looks like this:- 

SampleID,Tissue,Factor,Condition,Treatment,Replicate,bamReads,ControlID,bamControl,Peaks,PeakCaller

K1,White,EZH,Young,Treated,1,sorted_bam_backup/K1_NT_hssorted.bam,K7,sorted_bam_backup/K7_NT_hssorted.bam,peaks/macsNThsK1_peaks.bed,raw

K3,White,EZH,Young,Treated,2,sorted_bam_backup/K3_NT_hssorted.bam,K7,sorted_bam_backup/K7_NT_hssorted.bam,peaks/macsNThsK3_peaks.bed,raw

K5,White,EZH,Young,Treated,3,sorted_bam_backup/K5_NT_hssorted.bam,K7,sorted_bam_backup/K7_NT_hssorted.bam,peaks/macsNThsK5_peaks.bed,raw

The first four lines of my MACS bed file looks like this :- 

chr1    11003    11775    MACS_peak_1    210.13
chr1    28497    30226    MACS_peak_2    270.62
chr1    181304    182466    MACS_peak_3    232.53
chr1    198831    200680    MACS_peak_4    313.48

I have gone through various help pages already available on the internet but I am still unable to solve this problem. 

Can someone please help. 

Best Regards

Hemant Gujar. 

The peak files you are using are in bed format, with the score in the 5th column. In your sample sheet, you specified the format as "raw", which expects the score in the 4th column. So if you change the values in the last column of samples.csv from "raw" to "bed", it should work.

-Rory