Building from source

11 views
Skip to first unread message

Lachlan Simpson

unread,
Jan 19, 2015, 11:31:24 AM1/19/15
to genome...@soe.ucsc.edu
Hi,

I'm building Genome Browser from source. My role is one of a systems
administrator - the geneticist I'm installing this for gave me a one
hour run down on the bioscience and I mostly kept up, but still have
some questions from outside of my domain.

There are a number of READMEs within the code, but from what I can
gather, this is the canonical with regard to installation:

http://genome-source.cse.ucsc.edu/gitweb/?p=kent.git;a=blob;f=src/product/README.building.source

I was wondering if I could get some clarification with regard to the
term mirror. When referring to mirror in the code, are you referring
to running a local version of the UCSC genome browser? ie, with all
the data you have already uploaded and tagged?

Also, at the moment I've only been asked for "the genome browser" with
blat and blast search engines.

1. I note that there is both a blat and a webBlat dir in src - do I
want both? Is the blat a back end for webBlat?
2. I see no "blast" dir, or anything that looks like a more simple search
3. Where is the GB itself? Is that HG or product?

4. Finally, if you were a biologist, which of the apps in that src do
you use? Some seem useful (eg getgene), some seem non specific support
(lib, inc), others are opaque (proDust). Is there an easy way to use
all?

cheers
L.

Lachlan Simpson

unread,
Jan 19, 2015, 11:31:55 AM1/19/15
to genome...@soe.ucsc.edu
Also, following the instructions to the end, and installing on Ubuntu
14.04, I exectuted:

sudo make install DESTDIR=/var/www CGI_BIN=/usr/lib/cgi-bin
DOCUMENTROOT=/var/www/html

and it seems to have installed the cgi-bin files here?

/var/www/usr/lib/cgi-bin/

Restarting apache2 doesn't help - I still get the ubuntu default
apache page. I know I can change the ScriptAlias in the
sites-available conf to that new space, but I'd rather the cgi-bin
ended up in /usr/lib/cgi-bin, where other sysadmins would expect it.

I will look into it more closely in the morning.

cheers
L.

On 19 January 2015 at 15:55, Lachlan Simpson

Lachlan Simpson

unread,
Jan 20, 2015, 12:33:38 PM1/20/15
to genome...@soe.ucsc.edu
Both of these emails can be ignored. Further experimentation means I
have made some progress. Some issues from here in will be addressed in
another email.

On 19 January 2015 at 16:31, Lachlan Simpson
Reply all
Reply to author
Forward
0 new messages