Mapping assemblies using Blat

[Guy Naamati]

Hi genome browser (and Blat) people.
My name is Guy and I am a bioinformatician working on wheat.

I am now doing some work on a new wheat assembly, and one of the things I need to do is map the previous assembly to the new one.
I have been trying different tools (ATAC, lastz) with different levels of success, and was advised I should also try using Blat.

Wheat is a very big genome so my first test is trying to map one chrome from the old to the new.
However, this is going very very slowly, and I'm wondering if you have any ideas how to speed this up.

This is the command I am running (in the website it was advised to use this for Mapping DNA to DNA in the same species):
>blat old_ch1.fa new_ch1.fa -ooc=tgac.11.ooc -fastMap -out=blast8 out.blat.ch1.txt

The ooc file is created on the old assembly.
I also tried running it with different params without much success.

It is also important to add that the query file (new_ch1.fa) contains scaffold sequences which can be quite long (1000s of bp)

The problem is that for each of these queries I am getting on average over 1000 hits so the progress is very very slow.
So, the question: is there a way to run Blat efficiently for mapping assemblies, or is this just the wrong tool to use?

And if it is the correct tools, what would be the crucial param(s) I should change?

Thanks in advance
Guy Naamati (European Bioinformatics Institute)

[Hiram Clawson]

Good Morning Guy:

We call this the 'same species liftOver' procedure. The genomes are actually
broken up into 5,000 base chunks and each target chunk is aligned with blat against
each query chunk. The results of blat are then chained together. blat is not
designed for large query sequences, it can only manage 5,000 bases in a query.
Please note these sample scripts that mapped human hg19 assembly to human hg38:
http://genome-test.cse.ucsc.edu/~hiram/sameSpeciesLiftOver/
You can adjust these scripts for your specific circumstances. See also:
http://genomewiki.ucsc.edu/index.php/Same_species_lift_over_construction
The example script from that is a much more limited example:
http://genomewiki.ucsc.edu/images/9/91/SameSpeciesBlatSetup.sh.txt

--Hiram

On 2/8/16 6:32 AM, Guy Naamati wrote:
> Hi genome browser (and Blat) people.
> My name is Guy and I am a bioinformatician working on wheat.
>
> I am now doing some work on a new wheat assembly, and one of the things I need to do is map the
> previous assembly to the new one.
> I have been trying different tools (ATAC, lastz) with different levels of success, and was
> advised I should also try using Blat.
>
> Wheat is a very big genome so my first test is trying to map one chrome from the old to the new.
> However, this is going very very slowly, and I'm wondering if you have any ideas how to speed
> this up.
>
> This is the command I am running (in the website it was advised to use this for Mapping DNA to
> DNA in the same species):

> >blat *old_ch1.fa new_ch1.fa* -ooc=tgac.11.ooc -fastMap -out=blast8 out.blat.ch1.txt

[Guy Naamati]

Thank you for your answer