Hi, Yu!
I found a way that hopefully delivers what you want:
hgLoadChain -test hg18 galtChain
/gbdb/hg18/liftOver/hg18ToHg19.over.chain.gz
Normally we do not load the same-species chains like this.
-test causes it to create chain.tab in the current directory.
galtChain is just a dummy track name that is never used.
cat chain.tab | gawk '{print $3"\t"$5"\t"$6"\t"$9"\t"$7"\t"$10"\t"$11}'
> chain-minimal.tab
It's output is target (hg18) on the left, and query (hg19) on the right.
[hgwdev:~> head chain-minimal.tab
chr1 0 247199719 + chr1 10000 249233096
chr1 1609759 1609849 + chr1 1672838 1672928
chr1 2475290 2488984 - chr1 246751494 246765188
chr1 13192499 13192587 - chr1 236203315 236203403
[...]
The strand applies to the two sets.
If it is plus, nothing special happens.
If it is minus, you can reverse complement
either the query or the target, it's mathematically the same.
It is only 89K in size.
I have attached it to this email, which should reach you
but be stripped from the general list.
Note that these are the 0-based half-open coordinates
so that the first 100 bases of the chromosome are specified
as [0-100).
-Galt
On 11/09/11 20:46, Yuan Jian wrote:
> Hi
> UCSC provides liftover, but
> is it possible to get a table with two columns to show the different loci
> between hg18 and hg19:
> one column shows hg18 loci and the other is hg19 loci.
>
> thanks
> Yu
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
