how bwa and bowtie handle repeats?

henry · 11-26-2009, 01:55 AM

hg18 downloaded from ucsc consits of lower-case letters like a c g t.
human build 36 downloaded from ncbi consits of A C G T N.

Could anyone kindly tell me how bwa and bowtie deal with the Ns and lower-case letters when they map reads to the references?

Best

henry · 11-26-2009, 04:40 PM

Bwa replaces Ns in the human genome from NCBI build 36 with A,C,G,T randomly. Bowtie by default excludes Ns.

As for hg18 in UCSC, there are lower-case letters for representing repeat regions in the genome. How do bwa and bowtie handle them?
Bwa replaces those lower-case letters with A,C,G,T randomly, or changes lower-cases letters into upper-cases first?

Bowtie excludes those lower-cases letters or changes lower-case letters into upper-cases first?

Hope someone can tell me.

KevinLam · 02-03-2010, 12:18 AM

When gapped alignment is disabled, BWA is expected to generate the same alignment as Eland, the Illumina alignment program. However, as BWA change ‘N’ in the database sequence to random nucleotides, hits to these random sequences will also be counted. As a consequence, BWA may mark a unique hit as a repeat, if the random sequences happen to be identical to the sequences which should be unqiue in the database. This random behaviour will be avoided in future releases.

Hope this helps

nashp · 01-10-2011, 04:20 PM

Hey
I am also looking for the same answers. What does 'n' in a read mean? in a genome mean? how is it handled in each of them by bowtie?

Please let me know if you have found answers to these

- Nash

louis7781x · 01-20-2011, 11:30 PM

in bowtie ,use -m optioon ;for example, -m 2 means if output of alignment result >2 ,it's will print out "no alignment"

adamdeluca · 01-21-2011, 06:15 AM

Quote:

Originally Posted by nashp

Hey
I am also looking for the same answers. What does 'n' in a read mean? in a genome mean? how is it handled in each of them by bowtie?

Please let me know if you have found answers to these

- Nash

'n' means an ambiguous base call, an unknown. An N in a read is a mismatch, i.e. reported in the alignment as a T>N change for example. An N in the reference will cause the alignment to break (bowtie alignments can never span an N) the --ntoa option for building the index will convert all to A's so they can be treated as mismatches.

Liy · 04-30-2014, 12:52 AM

Hi all,

Can someone please tell me, how exactly handle bowtie2 the multiple aligned reads? I ran the following commad and the results showed that ~50% of my reads (10m) could align multiple times.

bowtie2 -x ChIP_IP1.fastq.gz -S ChIP1_local.sam --very-sensitive-local --no-unal -p 6

Any suggestion, what to do with multi hits?

Thanky you!

mastal · 04-30-2014, 03:52 AM

See the section of the Bowtie2 manual under the heading 'Reporting'

http://bowtie-bio.sourceforge.net/bo...html#reporting

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Bowtie 2 versus BWA	adaptivegenome	Bioinformatics	93	12-15-2013 04:15 PM
Bowtie vs BWA	sarbashis	Illumina/Solexa	12	08-26-2011 04:46 AM
how does bowtie handle n's in reads? ref sequence?	nashp	Bioinformatics	0	01-10-2011 04:09 PM
Main difference between Interspersed repeats and tandem repeats problem asking...	edge	General	0	06-17-2010 01:36 AM
Bowtie, BWA, strand information	Zimbobo	Bioinformatics	6	04-24-2010 08:30 AM

11-26-2009, 01:55 AM	#1
henry Member Location: china Join Date: Sep 2009 Posts: 36	how bwa and bowtie handle repeats? hg18 downloaded from ucsc consits of lower-case letters like a c g t. human build 36 downloaded from ncbi consits of A C G T N. Could anyone kindly tell me how bwa and bowtie deal with the Ns and lower-case letters when they map reads to the references? Best

01-10-2011, 04:20 PM	#4
nashp Junior Member Location: Calif, US Join Date: Jan 2011 Posts: 4	I m lookin for the same answers Hey I am also looking for the same answers. What does 'n' in a read mean? in a genome mean? how is it handled in each of them by bowtie? Please let me know if you have found answers to these - Nash

11-26-2009, 04:40 PM	#2
henry Member Location: china Join Date: Sep 2009 Posts: 36	Bwa replaces Ns in the human genome from NCBI build 36 with A,C,G,T randomly. Bowtie by default excludes Ns. As for hg18 in UCSC, there are lower-case letters for representing repeat regions in the genome. How do bwa and bowtie handle them? Bwa replaces those lower-case letters with A,C,G,T randomly, or changes lower-cases letters into upper-cases first? Bowtie excludes those lower-cases letters or changes lower-case letters into upper-cases first? Hope someone can tell me.

02-03-2010, 12:18 AM	#3
KevinLam Senior Member Location: SEA Join Date: Nov 2009 Posts: 203	When gapped alignment is disabled, BWA is expected to generate the same alignment as Eland, the Illumina alignment program. However, as BWA change ‘N’ in the database sequence to random nucleotides, hits to these random sequences will also be counted. As a consequence, BWA may mark a unique hit as a repeat, if the random sequences happen to be identical to the sequences which should be unqiue in the database. This random behaviour will be avoided in future releases. Hope this helps

01-20-2011, 11:30 PM	#5
louis7781x Member Location: Hong Kong Join Date: Oct 2010 Posts: 74	in bowtie ,use -m optioon ;for example, -m 2 means if output of alignment result >2 ,it's will print out "no alignment"

04-30-2014, 12:52 AM	#7
Liy Member Location: Hungary Join Date: Feb 2012 Posts: 19	Hi all, Can someone please tell me, how exactly handle bowtie2 the multiple aligned reads? I ran the following commad and the results showed that ~50% of my reads (10m) could align multiple times. bowtie2 -x ChIP_IP1.fastq.gz -S ChIP1_local.sam --very-sensitive-local --no-unal -p 6 Any suggestion, what to do with multi hits? Thanky you!

04-30-2014, 03:52 AM	#8
mastal Senior Member Location: uk Join Date: Mar 2009 Posts: 666	See the section of the Bowtie2 manual under the heading 'Reporting' http://bowtie-bio.sourceforge.net/bo...html#reporting