hg.conf parameters

19 views
Skip to first unread message

McCully, Dwayne (NIH/NIAMS) [C]

unread,
Dec 1, 2016, 5:46:28 PM12/1/16
to genome...@soe.ucsc.edu

The example hg.conf and hg.conf from the browserSetup.sh script have some differences.

 

Do the following items needs to be set?

 

# deactivate the hgMirror CGI on this machine

allowHgMirror=0

 

 

# use 2bit files instead of nib, this is only relevant in failover mode

# and for very old assemblies

forceTwoBit=yes

 

Dwayne

Maximilian Haeussler

unread,
Dec 1, 2016, 6:35:02 PM12/1/16
to McCully, Dwayne (NIH/NIAMS) [C], genome...@soe.ucsc.edu
Hi Dwayne,

It's safe to not worry about either setting.

I've updated the documentation in ex.hg.conf now. Both statements are in it in the new version.

allowHgMirror=0 is the default setting, so one can remove this line. It's only relevant on GBIB. I will remove this line from the script. Apparently it's confusing.

forceTwoBit=yes forces the browser to use 2bit files to get the DNA. This is faster over internet connections. It only makes sense for mirrors that load data remotely through the internet. The browserSetup.sh script will set your mirror to this mode, for a first round of testing, just to show that this works in principle but from the East coast and for most assemblies, you will see that this mode is too slow, so you probably will have to download your assemblies. The setting has no effect anymore once you have downloaded the assembly data with the other commands of the script, like "mirror".
This only affects these assemblies: dm1, cb1, hg16, and sacCer1 which I hope no one is using anymore.

Thanks for pointing this out!

Max

--


Maximilian Haeussler

unread,
Dec 2, 2016, 8:48:50 PM12/2/16
to McCully, Dwayne (NIH/NIAMS) [C], genome...@soe.ucsc.edu
Hi Dwayne,

thanks for going through our hg.conf options. It looks like we should do this ourselves from time to time. :-)

This option browser.indelOptions is apparently not needed anymore. We've now removed it from the code and from the documentation.

Don't hesitate to contact us if you find other options in need of removal.... :-)
Max

On Fri, Dec 2, 2016 at 9:35 AM, McCully, Dwayne (NIH/NIAMS) [C] <dmcc...@mail.nih.gov> wrote:
>
> Thanks Max for answering my question but I have one more.
>
>  
>
> What is the following used for?
>
>  
>
> # legacy setting
>
> browser.indelOptions=on
>
>  
>
> Dwayne

>
>  
>
> From: Maximilian Haeussler [mailto:m...@soe.ucsc.edu]
> Sent: Thursday, December 01, 2016 6:09 PM
> To: McCully, Dwayne (NIH/NIAMS) [C] <dmcc...@mail.nih.gov>
> Subject: Re: [genome-mirror] hg.conf parameters
>
>  
>
> allowHgMirror=0 is the default setting, so one can remove this line. It's only relevant on GBIB. I will remove this line from the script. Apparently it's confusing.
>
>  
>
> forceTwoBit=yes forces the browser to use 2bit files to get the DNA. This is faster over internet connections. It only makes sense for mirrors that load data remotely through the internet. The browserSetup.sh script will set your mirror to this mode, for a first round of testing, just to show that this works in principle but from the East coast and for most assemblies, you will see that this mode is too slow, so you probably will have to download your assemblies. The setting has no effect anymore once you have downloaded the assembly data with the other commands of the script, like "mirror".
>
>  That being said, I don't think that apart from 3-4 extremely old assemblies, we still have assemblies that  do not have a 2bit file, which is why I wrote this comment. This setting is only interesting for groups that still use an assembly from the very early 2000s, like ce2.
Reply all
Reply to author
Forward
0 new messages