maq map -n in colorspace

Mr Mutundes · 11-29-2009, 03:31 AM

Hi,

I have been investigating the use of MAQ 0.7.1 for aligning SOLiD reads to a reference and calling SNPs. I am wondering about the behaviour of the -n option which sets maximum mismatches allowed when using maq's map command. MAQ allows only values of 1,2 or 3 for this option. Since in colorspace a SNP is comprised of two adjacent color changes does this mean that reads with more than one SNP cannot be aligned with maq map?

Thanks for any help.

Mr M

Chipper · 11-29-2009, 06:36 AM

Not sure about MAQ, but bwa works well with CS. You can set it to allow 2 mismatches in the first (good part) of the read and several in the last part and still get good speed (and more unique alignments than mapreads). But my impression after some extensive testing is that BFAST is the best option since it is able to place more low-quality reads uniquely as well as using the first color. Have not yet tested if the low-quality alignments are correct though...

Mr Mutundes · 12-01-2009, 03:22 PM

Thanks Chipper

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
maq map problem	prm36	Bioinformatics	0	03-12-2010 10:12 AM
MAQ - colorspace alignment troubles	Jonathan	Bioinformatics	1	01-20-2010 11:16 PM
maq map command...	hcpark79	Bioinformatics	2	12-05-2009 06:29 AM
maq map -H file	baohua100	Bioinformatics	1	08-07-2009 09:06 AM
maq map/match settings in colorspace	der_eiskern	Bioinformatics	0	07-28-2009 02:10 PM

11-29-2009, 03:31 AM	#1
Mr Mutundes Member Location: Sydney, Australia Join Date: Jan 2009 Posts: 17	maq map -n in colorspace Hi, I have been investigating the use of MAQ 0.7.1 for aligning SOLiD reads to a reference and calling SNPs. I am wondering about the behaviour of the -n option which sets maximum mismatches allowed when using maq's map command. MAQ allows only values of 1,2 or 3 for this option. Since in colorspace a SNP is comprised of two adjacent color changes does this mean that reads with more than one SNP cannot be aligned with maq map? Thanks for any help. Mr M

11-29-2009, 06:36 AM	#2
Chipper Senior Member Location: Sweden Join Date: Mar 2008 Posts: 324	Not sure about MAQ, but bwa works well with CS. You can set it to allow 2 mismatches in the first (good part) of the read and several in the last part and still get good speed (and more unique alignments than mapreads). But my impression after some extensive testing is that BFAST is the best option since it is able to place more low-quality reads uniquely as well as using the first color. Have not yet tested if the low-quality alignments are correct though...

12-01-2009, 03:22 PM	#3
Mr Mutundes Member Location: Sydney, Australia Join Date: Jan 2009 Posts: 17	Thanks Chipper