-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathREADME.ensembl
168 lines (121 loc) · 6.16 KB
/
README.ensembl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
GRAMENE ENSEMBL; how to set up the web server
=============================================
This document describes how to configure the gramene-specific version
of the ensembl web browser. This server, generally proxied behind a
seperate gramene web server, provides the Gramene implementation of
Ensembl's genome browser. This readme corresponds to the 'new
version' which uses the ensembl-plugins system, rather than the
gramene_ensembl code base.
1. If you are installing on a machine that has Apache-2 installed, and
has run Ensembl in the past, skip to step 2. If not, follow the
instructions here;
http://www.ensembl.org/info/software/website/installation/non-ensembl-code.html
1a The install instructions are missing the 'emboss' suite, whose
installation is described here;
http://emboss.sourceforge.net/download
In addition, the emboss 'restrict' application needs to be set up
using 'rebaseextract';
http://emboss.sourceforge.net/apps/release/5.0/emboss/apps/rebaseextract.html
...download the 'withrefm' file and 'proto' files from the
following locations;
http://rebase.neb.com/rebase/link_withrefm
http://rebase.neb.com/rebase/link_proto
1b Install Microsoft TrueType fonts. For a Linux system, follow these
instructions:
http://corefonts.sourceforge.net/
The RPM will probably install the fonts into:
/usr/share/fonts/msttcorefonts
Ensembl, however, expects to find them in:
/usr/local/share/fonts/ttfonts
The latter needs to be a symlink to whatever the former ends up being.
2. Decide where to install the site. E.g. /usr/local.
$ cd /usr/local
3. Check out the Gramene project from the Gramene Subversion repository.
You may need to use a tmp dir, and then copy back into
/usr/local. The following example checks-out the build29-branch;
$ svn co http://svn.warelab.org/gramene/branches/build29-branch/ \
gramene-29
4. Check out the bioperl-live project from BioPerl's CVS repository.
You need to get the bioperl-release-1-2-3 branch. I find it
convenient to call the directory after the release;
$ cvs -d :pserver:[email protected]:/home/repository/bioperl \
co -r bioperl-release-1-2-3 -d bioperl-1-2-3 bioperl-live
(pass cvs)
5. Check out the Ensembl code from Ensembl's CVS. This is a little
more complex as several (aliased) projects are involved. Gramene 30
needs ensembl-55 and ensembl genomes (eg-plugins) 3-55.
$ mkdir ensembl-55
$ cd ensembl-55
$ cvs -d :pserver:[email protected]:/cvsroot/ensembl login
(pass CVSUSER)
$ cvs -d :pserver:[email protected]:/cvsroot/ensembl \
co -r branch-ensembl-55 ensembl-api ensembl-website
$ cvs -d :pserver:[email protected]:/cvsroot/ensembl \
co -r branch-ensemblgenomes-3-55 eg-plugins
$ cvs -d :pserver:[email protected]:/cvsroot/biomart login
(pass CVSUSER)
$ cvs -d :pserver:[email protected]:/cvsroot/biomart \
co -r release-0_6 biomart-perl
$ cd biomart-perl
$ <follow INSTALL> # There are some code mods to make as well...
$ cd ../
6. We now have all the code. Symlinks are currently required from the
ensembl root directory to the other projects above. I.e.
$ cd ensembl-55
$ ln -s ../gramene-30 gramene-live
$ ln -s ../bioperl-1-2-3 bioperl-live
7. A sylink is also required to the apache installation. If, for
example, apache2 is installed into /usr/local/apache2;
$ ln -s /usr/local/apache2 apache2
8. ensembl ships with an httpd configuration file;
ensembl-55/conf/httpd.conf. By default, the server will write its
logs to a logs directory, which needs creating;
$ mkdir ./logs
9. ensembl works using a plugin system. To use the Gramene-specific
plugins, copy the relevent Plugins.pm file into the ensembl conf
dir;
$ cp gramene-live/ensembl-plugins/gramene/conf/Plugins.pm conf/
10. Two main files are used to configure the web site;
- the 'default' ensembl file at conf/SiteDefs.pm and,
- the 'gramene' specific file at
gramene-live/ensembl-plugins/gramene/conf/SiteDefs.pm
Edit the gramene-specific file to ensure that the server starts on
the correct port, for example.
11. Assuming that the gramene SiteDefs.pm module has ENSEMBL_PORT set to
an unused port, and you have a mod-perl enabled httpd (apache2 for
ensembl versions > 42), you can now start the web server;
$ sudo /usr/local/apache2/bin/httpd -d /usr/local/apache2 \
-f $PWD/conf/httpd.conf
12.To ensure that the httpd starts on server startup, create an
apachectl start script, and configure the machine's /etc/init.d
system accordingly. By making the ensembl root dir a symlink, the
startup script can be somewhat generic;
$ cd ..
$ ln -s ensembl-55 ensembl-live
The httpd config file will then be
/usr/local/ensembl-live/conf/httpd.conf, and the pid file
will be /usr/local/ensembl-live/logs/httpd.pid
13.You should now be able to point a web browser at the port that you
started the server on, and see the site. E.g.
http://acorn.cshl.edu:8011/genome_browser/index.html
14.You may need to configure the files in
gramene-live/ensembl-plugins/gramene/conf/ini-files to point to the
correct (version 30-55) ensembl databases. This procedure probably
warrants a seperate README.
15.BioMart needs to be configured;
$ cd biomart-perl
$ perl bin/configure.pl -r conf/martRegistry.xml
$ perl sudo find conf/ -type d -exec chmod 777 {} \;
You need to rerun this step whenever the BioMart configuration is changed.
16.Add the following to root's crontab (sudo crontab -e):
# Clean ensembl* tmp and img-tmp files
0 4 * * * find /usr/local/ensembl*/*tmp -type f -atime +1 -exec rm {} \; > /dev/null 2>&1
0 4 * * * find /usr/local/ensembl*/img-cache -type f -atime +1 -exec rm {} \; > /dev/null 2>&1
Further info; read Ensembl's detailed installation docs;
http://www.ensembl.org/info/software/website/installation/index.html
To upgrade multi-species databases such as
ensembl_website_50
ensembl_web_user_db_50
download data from ftp://ftp.ensembl.org/pub/release-55/multi_species_50/data/mysql
and refer to doc at
http://www.ensembl.org/info/software/website/installation/build.html