Error while performing MACS program

Question

[Mohit]

He hi ,

I am getting some error when I perform MACS.

Step 1)For the alignment purpose I used the BWA alignment purpose, then i got .sam file.

Step 2) sam to bam conversion by samtools (View, sort, index these commands are used)

Step 3) .bam file is given input to the MACS for peak calling but getting some error....

INFO  @ Mon, 23 Jun 2014 11:08:02:  171000000
INFO  @ Mon, 23 Jun 2014 11:08:04: #1 tag size is determined as 51 bps
INFO  @ Mon, 23 Jun 2014 11:08:04: #1 tag size = 51
INFO  @ Mon, 23 Jun 2014 11:08:04: #1  total tags in treatment: 0
INFO  @ Mon, 23 Jun 2014 11:08:04: #1 user defined the maximum tags...
INFO  @ Mon, 23 Jun 2014 11:08:04: #1 filter out redundant tags at the same location and the same strand by allowing at most 1 tag(s)
INFO  @ Mon, 23 Jun 2014 11:08:04: #1  tags after filtering in treatment: 0
Traceback (most recent call last):
  File "/usr/local/bin/macs14", line 366, in <module>
    main()
  File "/usr/local/bin/macs14", line 83, in main
    info("#1  Redundant rate of treatment: %.2f" % (float(t0-t1)/t0))
ZeroDivisionError: float division by zero

[Mohit]

@Fedor Gusev:

I have used -t only , I don't have control!!

Can u tell me, "Is there any thing like for SoLiD data we shouldn't use the BWA algorithm for Alignment?" Because I checked at the sam file obtained from BWA output, It do not contains sequence. It contains only the NNNNNNNN..... so I wondered!

Thanking You for consideration and time!!

Mohit R. Navandar

Answer 1

[Fedor Gusev]

BWA does not support colorspace reads since version 0.6.1. I assume you input file is incorrect; you have to use another aligner.

I suspect that you provided -c option for MACS instead of -t. ChIP-seq experiments usually (but no necessary) have two experiments: treatment and control. If you don't have a control, then you should provide only -t option.