Tool for Human to Gorilla alignment

Avazeh · 08-20-2012, 04:47 AM

Hi all,

I have 400 human seqs to align to Gorilla and Chimp. Tried UCSC and got a useful alignment for Chimp but the Gorilla one has supercontigs rather than chromosome numbers

I have to add 2kb to the aligned seqs and then pass to another tool. But his is not possible with just having supercontigs numbers! So I am looking for an alternative.

Any idea or thought would be much appreciated,
Avazeh

ishmael · 08-20-2012, 07:01 AM

Would you mind to provide more details?
What's the tool you used to do the aignment? liftover?
What's the nature of your data?

Avazeh · 08-20-2012, 07:15 AM

Sure:

I have 400 human seqs, each between 2kb to 4kb. I need to map them to Gorilla and Chimp and then look for motifs in the extended alignments found. Submitted the regions to UCSC table browser and downloaded conservation in Comparative Genome group, using Multiz Align option to get the sequences in the alignments too. Unfortunately these alignment are just available on gorGOR1 which is the first Gorilla build and contains contig number rather than chromosome number! So I really don't know where on Gorilla genome they hit and cannot extend it to do motif analysis

I need to align my regions to Gorilla's 2011 built, gorGOR3, in which the supercontigs are resolved into chromosomes.

Any help is appreciated

dpryan · 08-20-2012, 07:27 AM

Quote:

Originally Posted by Avazeh

I need to align my regions to Gorilla's 2011 built, gorGOR3, in which the supercontigs are resolved into chromosomes.

gorGOR3 is available from Ensembl, so I would recommend just downloading the fasta files and aligning to them.

ishmael · 08-20-2012, 07:28 AM

If all sequences are form human, I would suggest use reciprocal liftover approach.
1. download liftover program from UCSC.
2. download liftover chain files: both of human to gorilla and gorilla to human. You can find them under the genome file download pages.
3. mapping the sequences to human genome, and get the coordination in bed format.
4. convert human bed to gorilla genome by liftover, allow multiple convertion.
5. convert the converted gorilla bed file back to human genome, also multiple loci allowed.
6. filtered the results that the gorilla and human could be converted reciprocally.
That's the protocol I think relatively reliable.

Avazeh · 08-21-2012, 01:47 AM

Thank you both, will try them out

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Alignment tool for long reads?	naragam	Bioinformatics	15	04-05-2013 05:53 PM
Tool for analyzing low-coverage human data	pravee1216	Bioinformatics	0	05-17-2012 07:11 PM
human genome build identifier tool	rworthi	Bioinformatics	0	06-18-2011 08:44 AM
Alignment tool for use with ambiguous reference?	lindseyjane	Bioinformatics	13	11-18-2010 10:44 AM
The Best Tool for RNA Alignment	EthoStrain	Bioinformatics	5	09-07-2010 06:56 AM

08-20-2012, 04:47 AM	#1
Avazeh Junior Member Location: Bath Join Date: Jul 2012 Posts: 7	Tool for Human to Gorilla alignment Hi all, I have 400 human seqs to align to Gorilla and Chimp. Tried UCSC and got a useful alignment for Chimp but the Gorilla one has supercontigs rather than chromosome numbers I have to add 2kb to the aligned seqs and then pass to another tool. But his is not possible with just having supercontigs numbers! So I am looking for an alternative. Any idea or thought would be much appreciated, Avazeh

08-20-2012, 07:01 AM	#2
ishmael Member Location: NY, US Join Date: Jul 2008 Posts: 17	Would you mind to provide more details? What's the tool you used to do the aignment? liftover? What's the nature of your data?

08-20-2012, 07:15 AM	#3
Avazeh Junior Member Location: Bath Join Date: Jul 2012 Posts: 7	Sure: I have 400 human seqs, each between 2kb to 4kb. I need to map them to Gorilla and Chimp and then look for motifs in the extended alignments found. Submitted the regions to UCSC table browser and downloaded conservation in Comparative Genome group, using Multiz Align option to get the sequences in the alignments too. Unfortunately these alignment are just available on gorGOR1 which is the first Gorilla build and contains contig number rather than chromosome number! So I really don't know where on Gorilla genome they hit and cannot extend it to do motif analysis I need to align my regions to Gorilla's 2011 built, gorGOR3, in which the supercontigs are resolved into chromosomes. Any help is appreciated

08-20-2012, 07:28 AM	#5
ishmael Member Location: NY, US Join Date: Jul 2008 Posts: 17	If all sequences are form human, I would suggest use reciprocal liftover approach. 1. download liftover program from UCSC. 2. download liftover chain files: both of human to gorilla and gorilla to human. You can find them under the genome file download pages. 3. mapping the sequences to human genome, and get the coordination in bed format. 4. convert human bed to gorilla genome by liftover, allow multiple convertion. 5. convert the converted gorilla bed file back to human genome, also multiple loci allowed. 6. filtered the results that the gorilla and human could be converted reciprocally. That's the protocol I think relatively reliable.

08-21-2012, 01:47 AM	#6
Avazeh Junior Member Location: Bath Join Date: Jul 2012 Posts: 7	Thank you both, will try them out