de novo contigs don't match mapped reads

bdorsey · 09-24-2015, 02:05 PM

Hi all,

I'm confused about the results I get when I map the reads back to the contigs from Abyss. There are a number of contigs where the majority of reads have a different base than the contig at a given position. I don't understand how the assembler would assign a base to a position if it is not supported by the reads.

Thanks for any explanations.

Cheers,
BD

SylvainL · 09-25-2015, 12:39 AM

Hi,

maybe you have more reads supporting the base at this position than reads which don't support it? Different possibilities to check this: the fastest (if you already have a pipeline for) would be to do a fast SNP calling using your contigs as reference...

s.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DESeq2 - COLUMN and ROWS number don't match	KYR	RNA Sequencing	1	07-16-2014 02:21 PM
Too little reads mapped back to contigs using Bowtie	oliviaxinw	Bioinformatics	4	09-28-2012 06:18 AM
Can bowtie align mates individually if they don't match?	history_of_robots	Bioinformatics	1	02-06-2012 06:01 AM
How to allow mapped contigs to be used in de novo assembly projects	edleloth	454 Pyrosequencing	1	03-08-2011 08:13 AM
454IsotigsLayout.txt - contigs don't match to any isotig?	dschika	Bioinformatics	9	10-11-2010 05:58 AM

09-24-2015, 02:05 PM	#1
bdorsey Junior Member Location: USA Join Date: Sep 2015 Posts: 1	de novo contigs don't match mapped reads Hi all, I'm confused about the results I get when I map the reads back to the contigs from Abyss. There are a number of contigs where the majority of reads have a different base than the contig at a given position. I don't understand how the assembler would assign a base to a position if it is not supported by the reads. Thanks for any explanations. Cheers, BD

09-25-2015, 12:39 AM	#2
SylvainL Senior Member Location: Geneva Join Date: Feb 2012 Posts: 179	Hi, maybe you have more reads supporting the base at this position than reads which don't support it? Different possibilities to check this: the fastest (if you already have a pipeline for) would be to do a fast SNP calling using your contigs as reference... s.