Can Illumina reads be mapped to a different species ?

emorin · 01-25-2011, 09:32 AM

Hello,

I'm working on fungi genomes and I was wondering if I could map Illumina reads from one species (A) to gene models from an other one (B)?
I have tried with maq with -n (Number of maximum mismatches) set to 9 for 63 bp reads and only 50 000 reads on 25 millions mapped to the gene models. I know that I have to consider mapping in non coding regions but still the number is very low.
Am I trying to do something dummy ?

The goal is to try to generate a set of gene models for A from B. While waiting for the genome assembly.

Thanks,

Emmanuelle

doc.ramses · 01-26-2011, 02:02 AM

I think it's worth a try. What about not mapping against the whole genome but only the conserved regions ? Don't know if this is a good idea or if you have the needed data for this. It's just a lucky guess, but maybe I can point you to the right direction.

colindaven · 01-26-2011, 04:34 AM

9 bp seems like a lot of error in a 63 bp read. Do you know how divergent the two genesets are ? Have you tried Blastn to test the divergence ?
How about BLASTn for 1000 reads to geneset B ?

Another approach might be to try de novo assembly of the reads, for example with Velvet, and then map them to geneset B.

emorin · 01-26-2011, 05:04 AM

Hello,

Thanks for the answers.
A liltle bit more information about my data, they belong to 2 fungi of the same genus so pretty close.

To answer Doc ramses, I try to map to gene models that should be conserved enough. What do you mean by conserved regions ?

To colindaven, I agree with you about 9 being a lot, that why I posted my question. Apparently we assumed the 2 genesets are 15% divergent at the amino acid level. The assembly step before the mapping could be an idea.

I will proceed blastn analysis and let you know.

Thanks,

Emmanuelle

krobison · 01-26-2011, 07:21 AM

From the category Comparative Genomics in the Software Wiki:

LAST "Furthermore, when mapping Drosophila melanogaster reads to the Drosophila simulans genome, it increased the amount of correctly mapped reads from 49 to 66%"

DIAL "A computational pipeline for identifying single-base substitutions between two closely related genomes without the help of a reference genome."

I have not used either one.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Only 2% mapped reads?????	Jeannine	Bioinformatics	2	01-25-2012 06:20 PM
Error during PicardMarkDuplicates (Illumina paired-ends mapped by Bfast)	david.tamborero	Bioinformatics	2	08-09-2011 02:08 AM
mapped reads or mappable reads for normalization??	dzavallo	RNA Sequencing	0	05-19-2011 12:55 PM
Paired-end reads mapped 50% while each pair reads mapped 80%	zack80.liu	Bioinformatics	3	03-03-2011 02:06 AM
Combined mapping of RNA-Seq reads originating from multiple species	schelhorn	RNA Sequencing	7	11-05-2010 09:55 AM

01-25-2011, 09:32 AM	#1
emorin Junior Member Location: FRANCE Join Date: Sep 2010 Posts: 2	Can Illumina reads be mapped to a different species ? Hello, I'm working on fungi genomes and I was wondering if I could map Illumina reads from one species (A) to gene models from an other one (B)? I have tried with maq with -n (Number of maximum mismatches) set to 9 for 63 bp reads and only 50 000 reads on 25 millions mapped to the gene models. I know that I have to consider mapping in non coding regions but still the number is very low. Am I trying to do something dummy ? The goal is to try to generate a set of gene models for A from B. While waiting for the genome assembly. Thanks, Emmanuelle

01-26-2011, 02:02 AM	#2
doc.ramses Member Location: Planet Earth Join Date: Jan 2011 Posts: 26	I think it's worth a try. What about not mapping against the whole genome but only the conserved regions ? Don't know if this is a good idea or if you have the needed data for this. It's just a lucky guess, but maybe I can point you to the right direction.

01-26-2011, 04:34 AM	#3
colindaven Senior Member Location: Germany Join Date: Oct 2008 Posts: 415	9 bp seems like a lot of error in a 63 bp read. Do you know how divergent the two genesets are ? Have you tried Blastn to test the divergence ? How about BLASTn for 1000 reads to geneset B ? Another approach might be to try de novo assembly of the reads, for example with Velvet, and then map them to geneset B.

01-26-2011, 05:04 AM	#4
emorin Junior Member Location: FRANCE Join Date: Sep 2010 Posts: 2	Hello, Thanks for the answers. A liltle bit more information about my data, they belong to 2 fungi of the same genus so pretty close. To answer Doc ramses, I try to map to gene models that should be conserved enough. What do you mean by conserved regions ? To colindaven, I agree with you about 9 being a lot, that why I posted my question. Apparently we assumed the 2 genesets are 15% divergent at the amino acid level. The assembly step before the mapping could be an idea. I will proceed blastn analysis and let you know. Thanks, Emmanuelle

01-26-2011, 07:21 AM	#5
krobison Senior Member Location: Boston area Join Date: Nov 2007 Posts: 747	From the category Comparative Genomics in the Software Wiki: LAST "Furthermore, when mapping Drosophila melanogaster reads to the Drosophila simulans genome, it increased the amount of correctly mapped reads from 49 to 66%" DIAL "A computational pipeline for identifying single-base substitutions between two closely related genomes without the help of a reference genome." I have not used either one.