MAQ Help

[dafil]

Hi, I am trying out mapping few million short reads(>5X coverage) using MAQ. I am trying to map the reads against hg18 ref genome. Now, the problem I face is that it gives me a lot of 'n' s in most part of my consensus sequence.
Something like
>chr1
aacgatagatacagatacACGATGCATGACagcatgcatgacnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnaagcaGTCAGTACGATCGACTAGCAaagctagatannnnnnnnnnnnnnnnnn

what should be the right command for MAQ to map amd asseble to get rid of the 'n's?

thanks

[francois.sabot]

Hi,
Based on my personal experience, with only a 5x you will not be able to get off those nnn... They represent the non-resequenced regions, as 5x is a mean coverage... Some regions are highly overrepresented, some not at all sequenced...

[dafil]

thanks for your reply. i suspected that, but now i am pretty sure that is what happened, coverage might the issue here.

[jgibbons1]

Definitely a coverage issue. I am doing similar analysis right now and am getting the same thing with a mean coverage of about 11x. You may want to bring the alignment into a viewer to see for yourself where the low coverage islands and high coverage mountains are.