BWA and masking sequences...

spindrift · 02-17-2010, 02:57 AM

I have a set of data that has a linker (not barcode) that needs to be masked/removed before aligning to the genome. My aligner of choice is BWA.

If I mask these linkers (using some dynamic programming...) is it possible to use BWA to align these sequences? That is can I replace the linker with 'N's or similar characters and then align? I've tried replacing with 'N's but then the sequences which contain linkers aren't aligned as they then have too many mismatches.

After no joy with the above I've gone for the option of removing the linker sequences, giving me a mixed set of read lengths which I have to align separately and then recombine the SAM files to create a final BAM file.

Is this the best way of tackling this problem or am I missing something obvious?

Thanks

lh3 · 02-17-2010, 04:54 AM

Bwa works when you have mixed lengths of reads. Maq does not.

spindrift · 02-17-2010, 06:30 AM

Thanks for you prompt response, that seems like a good solution.

Thomas Doktor · 02-17-2010, 09:51 AM

The FASTX Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/) has a utility called fastx_clipper which might be able to remove your linker sequences.

spindrift · 02-17-2010, 11:20 AM

Thanks, that looks very useful indeed.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
BWA - number of sequences aligned	Cbon	Bioinformatics	7	05-13-2012 04:45 PM
BWA on raw sequences?	history_of_robots	Bioinformatics	3	03-12-2012 09:22 AM
BWA & Reverse Sequences	dp05yk	Bioinformatics	1	06-14-2011 08:15 PM
BWA: reference sequences longer than 4GB	ElMichael	Bioinformatics	1	06-06-2011 01:22 PM
Exome caputure: masking the non-genic sequences before aligning ?	lindenb	Bioinformatics	1	12-17-2010 07:10 AM

02-17-2010, 02:57 AM	#1
spindrift Junior Member Location: UK Join Date: Jun 2009 Posts: 3	BWA and masking sequences... I have a set of data that has a linker (not barcode) that needs to be masked/removed before aligning to the genome. My aligner of choice is BWA. If I mask these linkers (using some dynamic programming...) is it possible to use BWA to align these sequences? That is can I replace the linker with 'N's or similar characters and then align? I've tried replacing with 'N's but then the sequences which contain linkers aren't aligned as they then have too many mismatches. After no joy with the above I've gone for the option of removing the linker sequences, giving me a mixed set of read lengths which I have to align separately and then recombine the SAM files to create a final BAM file. Is this the best way of tackling this problem or am I missing something obvious? Thanks

02-17-2010, 04:54 AM	#2
lh3 Senior Member Location: Boston Join Date: Feb 2008 Posts: 693	Bwa works when you have mixed lengths of reads. Maq does not.

02-17-2010, 06:30 AM	#3
spindrift Junior Member Location: UK Join Date: Jun 2009 Posts: 3	Thanks for you prompt response, that seems like a good solution.

02-17-2010, 09:51 AM	#4
Thomas Doktor Senior Member Location: University of Southern Denmark (SDU), Denmark Join Date: Apr 2009 Posts: 105	The FASTX Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/) has a utility called fastx_clipper which might be able to remove your linker sequences.

02-17-2010, 11:20 AM	#5
spindrift Junior Member Location: UK Join Date: Jun 2009 Posts: 3	Thanks, that looks very useful indeed.