Installing GBrowse2 in your Home Directory

OK. This isn’t as hard as most people seem to blather on about. The key is understanding. If you don’t want to grok Perl, you should be talking to your sysadmin right now. Go on… click back right now and go get help!

First, install ‘local::lib’ [1] (assuming you haven’t done that already). It is important that you understand this step in all it’s Perlish detail, otherwise you’re not going to understand anything that follows. We’re going to be relying heavily on the ‘cpan’ installer (as a non-root user). The magical local::lib is the Perl module designed to let us do that! Before installing local::lib, You may want to scratch any ~/.cpan or ~/perl5 stuff you have half-heartedly installed to date. The magical local::lib is going to help you do that right.

Once you have local::lib correctly installed…

I believe there must be a trick to configuring cpan at this point that is left out of this narrative. It is probably in the lib::local docs. Scott 18:47, 3 June 2011 (UTC)

Test that cpan is working for you as a non-root user (install a random module or two).

Second, install BioPerl via cpan [2]. This can be painful. The best trick is to install all the dependencies first, then install BioPerl.

Third, follow the GBrowse install instructions (using cpan again)[3]. It should be easy to grab any prerequisites (using cpan).

Finally, the trick is to specify ‘user’ locations for all the directories that GBrowse expects: Here are some notes I made to help me keep track of that:

conf for example /etc/gbrowse2

htdocs for example /var/www/html/gbrowse2

tmp for example /var/tmp/gbrowse2

databases for example /var/www/html/gbrowse2/databases

cgibin for example /var/www/cgi-bin/gb2

modules for example /etc/httpd/modules
/etc/httpd/modules # /homes/dan/httpd/modules

wwwuser for example www-data

I forgot to mention, installing Apache is beyond the scope of this HOWTO.