[Genome] liftOver application conversion error

[Galt Barber]

Hi, Tim.

Please write your questions to genome at soe.ucsc.edu mailing list.
You should also sign up to the list,
and there are also online help archives one can search and browse,
see the FAQs.

I assume that you are using the online cgi hgLiftOver,
and not the standalone commandline tool liftOver{.exe} ?

What are you lifting, how many columns,
and from what assembly to what assembly?

There are three possibilities I can think of
for the start being reduced by 1:

1. We either input/output BED or Position with liftover.
BED files are in half-open ucsc 0-based coordinates [start,end)
Position input is meant for human rather than computer,
so that users can paste in the range from e.g. the browser position/search box.
These are in closed 1-based coordinates [start,end]

2. This is appears to happen, but only when the - strand in the
chain flips it over.  After further examination, one realizes that
this is actually correct.

3. There is a bug in liftOver.

Since many people use the tool and it hasn't changed much lately,
I am inclined to think #3 unlikely, although it can't be ruled out yet.

If one takes a browser position,
and tries to re-format it as a simple bed (e.g. bed3 or bed4),
and does not remember to subtract one from the start,
and then runs it through BED liftover input/output,
one may be surprised and confused.

If you are still having trouble, please send
a simple example.  (the mailing list does not accept attachments however).


-Galt


On Fri, 9 May 2008, Branwyn Wagman wrote:

> A message for the browser group.
>
> Branwyn
>
> -------- Original Message --------
> Subject: 	liftOver application conversion error
> Date: 	Fri, 9 May 2008 14:31:48 +0000
> From: 	TIM A. HINSLEY <thinsley at hotmail.com>
> To: 	<cbseweb at cbse.ucsc.edu>
>
>
>
> Hi
>
> I've just been using your liftOver conversion tool and noticed that its 
> altered some of my sequence positions incorrectly.
>
> All of the Bed files (original, unmapped and mapped files) produced by 
> liftOver have incorrectly deducted 1 bp from the start position of my 
> sequences. Though the other .txt file (produced as I was using the -positions 
> option) produced by liftOver was fine.
>
> Tim
>
> ------------------------------------------------------------------------
> Get 5GB of online storage for free! Get it Now! 
> <http://clk.atdmt.com/UKM/go/msnnkmgl0010000005ukm/direct/01/>
>
> -- 
> Branwyn Stewart Wagman Manager, Communications & Human Resources Center for 
> Biomolecular Science and Engineering (CBSE) Institute for Quantitative 
> Biomedical Research (QB3) 
> 501C Engineering 2 Building UC Santa Cruz 1156 High Street, MS: CBSE/ITI
> Santa Cruz CA 95064 Tel: (831) 459-3077
> Fax: (831) 459-1809
> bwagman at soe.ucsc.edu http://www.cbse.ucsc.edu
>
>