bowtie sam output, number of mismatches

sridharacharya · 01-07-2011, 03:15 PM

Hi,

If I am right, bowtie options -n 0 -l 16 should not allow any mismatches in the seed region of 16. But the following result is puzzling me.

I am aligning sequencing reads using the above options to miRNA sequences.
In SAM format output, I can find some lines with mismatches right in the beginning. Why is this?

I suppose the field containing "MD:" denotes the positions of mismatches. For the first record, there is a mismatch in the first position and for the second one, there is mismatch on the 26th position. For the above options, bowtie should not report mismatches in the seed region, then why is this happenning. Or Am I reading it wrongly?

2_1977_1917_F3 16 106b 23 255 48M * 0 0 GCGTTTCCGGACTACATCGGCGAACCTGATCTGCACTGTCAGCACTTT qqqqqqqqqqqq!!qqqqqqqqq!!qqqqqq!!qqqqqqqqqqqqqqq XA:i:0 MD:Z:0A1A0G1G1G1T1G0C2G0G0A1A0G2C1A0C0T20 NM:i:17 CM:i:27
2_1977_1040_F3 0 7-3 32 255 48M * 0 0 GGAAGACTAGTGATTTTGTTGTTCTAAGGTTTTACGACTACAATTCCC qqqqqqqqqqqqqqqqqqqqqqq!!qqqqqqqqqq!!qqq!!qqqqqq XA:i:0 MD:Z:25G1T2A0C6A4G2A1 NM:i:7 CM:i:16

frozenlyse · 01-08-2011, 06:17 PM

The flag of the first read (field 2) is 16 (0x10) which means that the read mapped to the reverse strand - the SAM format specifies that all alignments are to be represented on the forward strand, so the read is reported as the reverse complement, so the mismatches you are seeing are actually at the 3' end of the read

sridharacharya · 01-08-2011, 06:22 PM

Thanks frozenlyse. Clears my doubt!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error in bowtie's sam output?	burt	Bioinformatics	11	11-25-2012 04:53 PM
BWA:getting hits with given number of mismatches and indels	Chandana	Bioinformatics	0	01-11-2012 11:03 AM
Number of mismatches in Bowtie	Rachelly	Bioinformatics	0	05-08-2011 01:55 AM
sam to Bowtie output	seq_GA	Bioinformatics	0	03-21-2011 11:56 PM
Bowtie and SAM output	rdeborja	Bioinformatics	0	12-03-2009 11:30 AM

01-07-2011, 03:15 PM	#1
sridharacharya Member Location: Institute, WV Join Date: May 2010 Posts: 24	bowtie sam output, number of mismatches Hi, If I am right, bowtie options -n 0 -l 16 should not allow any mismatches in the seed region of 16. But the following result is puzzling me. I am aligning sequencing reads using the above options to miRNA sequences. In SAM format output, I can find some lines with mismatches right in the beginning. Why is this? I suppose the field containing "MD:" denotes the positions of mismatches. For the first record, there is a mismatch in the first position and for the second one, there is mismatch on the 26th position. For the above options, bowtie should not report mismatches in the seed region, then why is this happenning. Or Am I reading it wrongly? 2_1977_1917_F3 16 106b 23 255 48M * 0 0 GCGTTTCCGGACTACATCGGCGAACCTGATCTGCACTGTCAGCACTTT qqqqqqqqqqqq!!qqqqqqqqq!!qqqqqq!!qqqqqqqqqqqqqqq XA:i:0 MD:Z:0A1A0G1G1G1T1G0C2G0G0A1A0G2C1A0C0T20 NM:i:17 CM:i:27 2_1977_1040_F3 0 7-3 32 255 48M * 0 0 GGAAGACTAGTGATTTTGTTGTTCTAAGGTTTTACGACTACAATTCCC qqqqqqqqqqqqqqqqqqqqqqq!!qqqqqqqqqq!!qqq!!qqqqqq XA:i:0 MD:Z:25G1T2A0C6A4G2A1 NM:i:7 CM:i:16

01-08-2011, 06:17 PM	#2
frozenlyse Senior Member Location: Australia Join Date: Sep 2008 Posts: 136	The flag of the first read (field 2) is 16 (0x10) which means that the read mapped to the reverse strand - the SAM format specifies that all alignments are to be represented on the forward strand, so the read is reported as the reverse complement, so the mismatches you are seeing are actually at the 3' end of the read

01-08-2011, 06:22 PM	#3
sridharacharya Member Location: Institute, WV Join Date: May 2010 Posts: 24	Thanks frozenlyse. Clears my doubt!