博文
UCSC Genome Browser 本地化 on Redhat
||
Installing a Browser Mirror on Linux Red Hat
#环境：Red Hat Enterprise Linux Server release 6.2 (Santiago)；Linux version 2.6.32-220.el6.x86_64
#安装Mysql数据库。
yum -y install mysql-server
#新建一个用户。mysql –u 用户名 –p 密码
create user 'user'@'localhost' identified by 'PASSWORD';
grant all privileges on hg19.* to 'user'@'localhost'
flush privileges;
#安装Apache。
yum -y install httpd
sudo service apache2 restart
sudo yum install libssl0.9.8
sudo ln -s /usr/lib/libssl.so.0.9.8 /usr/lib/libssl.so.6
sudo ln -s /usr/lib/libcrypto.so.0.9.8 /usr/lib/libcrypto.so.6
sudo mkdir /var/www/genomebrowser
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/htdocs/ /var/www/genomebrowser/
sudo mkdir -p /var/www/genomebrowser/cgi-bin
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/cgi-bin/ /var/www/genomebrowser/cgi-bin/
sudo chown -R 755 /var/www/genomebrowser/cgi-bin/
#给genomebrowser访问权限
sudo chown -R 755 /var/www/genomebrowser/
#建立缓存文件夹
sudo rm /var/www/genomebrowser/trash
sudo mkdir /var/www/genomebrowser/trash
sudo chown -R 755 /var/www/genomebrowser/trash
#提供Javascript文件
sudo mkdir -p /usr/local/apache/htdocs/
sudo ln -s /var/www/genomebrowser/js/ /usr/local/apache/htdocs/js
sudo ln -s /var/www/genomebrowser/style/ /usr/local/apache/htdocs/style
#可以基于官网提供的.conf进行修改，官网的版本还有很多额外的功能，下载地址如下：
http://genome-test.cse.ucsc.edu/~kent/src/unzipped/product/ex.hg.conf
gedit /var/www/genomebrowser/cgi-bin/hg.conf
#添加以下内容：
# Configuration file for the UCSC Human Genome server
defaultGenome=Human
browser.documentRoot=/var/www/genomebrowser
# the format is in the form of name/value pairs, written as 'name=value'
# note that there is no space between the name and its value. Also, no blank lines #should be in this file.
# db.host is the name of the MySQL host to connect to
db.host=localhost
# db.user is the username used when connecting to the host
db.user=user
# this is the password to use with the above hostname
db.password=PASSWORD
db.trackDb=trackDb
# central.host is the name of the host of the central MySQL
# database where stuff common to all versions of the genome
# and the user database is stored.
central.db=hgcentral
central.host=localhost
central.user=user
central.password=PASSWORD
central.domain=
backupcentral.db=hgcentral
backupcentral.host=localhost
backupcentral.user=user
backupcentral.password=PASSWORD
backupcentral.domain=
sudo chown -R 755 /var/www/genomebrowser/cgi-bin/hg.conf
#配置Apache服务器，使下载的html、cgi-bin可以正常显示。#进入/usr/local/apache/conf目录。httpd.conf文件是Redhat系统中Apache的主要配置文件，谨慎添加内容，用文本编辑器建立httpd.conf并添加如下内容。位置可以放在<Directory "/var/www/cgi-bin">...</Directory>下一行。
gedit /usr/local/apache/conf/httpd.conf
###########################以下为添加的内容#####################Setting for genomebrowser in httpd.conf
#Enable SSI
XBitHack on
<Directory /var/www/genomebrowser/>
AllowOverride AuthConfig
Options Indexes FollowSymLinks MultiViews Includes
Order allow,deny
allow from all
AddType text/html .shtml
AddOutputFilter INCLUDES .shtml
</Directory>
#Enable CGI
ScriptAlias /genomebrowser/cgi-bin/ "/var/www/genomebrowser/cgi-bin/"
<Directory "/var/www/genomebrowser/cgi-bin">
AllowOverride None
Options +ExecCGI -MultiViews +SymLinksIfOwnerMatch
Order allow,deny
Allow from all
AddHandler cgi-script cgi pl
</Directory>
#进行到此步应该已经可以用http://localhost/genomebrowser/访问本地化的UCSC，如果显示不正常，则可能是由于Server Side Includes (SSI) on Apache没有正常启动
#重新检查以上/usr/local/apache/conf/httpd.conf是否配置正确，按照以上说明进行正确配置即可。
#重启apache2服务器
sudo service apache2 restart
#若此时出现CGI "Internal Server Error" 错误，则可以通过更改/var/www/genomebrowser/cgi-bin/的权限来解决chmod 755 -R /var/www/genomebrowser/cgi-bin
sudo service apache2 restart
#Can't start query: INSERT userDb VALUES(0,"",0,now(),now(),0)
#mySQL error 1036: Table 'userDb' is read only
#可以通过修改/var/lib/mysql的权限来解决
sudo chown -R mysql:mysql /var/lib/mysql
#重启mysqlsudo service mysql restart
#此时可以看到genomebrowser，如果显示不正常，导航栏异常，原因在于Redhat下Apache没有默认地读取htdocs（Ubuntu下Apache默认是读htdocs的）#解决这个问题只需将genomebrowser下的js和style两个文件夹复制到www下即可。
cp -R /var/www/genomebrowser/style /var/www/
cp -R /var/www/genomebrowser/js /var/www/
chmod 777 -R /var/www/genomebrowser/trash/
#如果遇到导航栏中某些链接指向www/cgi-bin的问题（因为默认的UCSC浏览器是放在www下的），目前比较保守的解决办法是将genomebrowser/cgi-bin下的文件复制到www/cgi-bin(注意此方法并未完全解决链接问题)cp -Rf /var/www/genomebrowser/cgi-bin/* /var/www/cgi-bin/
#建立Mysql tables。#首先要下载hgcentral.sql，把他放进mysql的目录，然后写入本地数据库文件。
wget http://hgdownload.cse.ucsc.edu/admin/hgcentral.sql
mysql -u root -p -e "create database hgcentral"
mysql -u root -p hgcentral < /home/download/hgcentral.sql
mysql -u root -p -e "grant all privileges on hgcentral.* to 'user'@'localhost'"
mysql -u root -p -e "create database hgFixed"
mysql -u root -p -e "grant select on hgFixed.* to 'user'@'localhost'"
#以hg19为例，如果要整个下载hg19的数据库，可以用下面的命令，请务必保证硬盘在500G以上（网速好的话，尚需3-4天下载完所有数据）
sudo mkdir /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/ /var/lib/mysql/hg19
#根据吕杰（http://blog.csdn.net/biocq/article/details/8954749）的经验，hg19数据库中至少要有以下几个表
grp
trackDb
hgFindSpec
chromInfo
gold
gap
#因此分别利用以下语句下载这些表（注，每个表都对应下面的三个文件，.MYD .MYI和.frm）
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/grp.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/grp.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/grp.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/trackDb.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/trackDb.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/trackDb.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/hgFindSpec.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/hgFindSpec.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/hgFindSpec.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/chromInfo.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/chromInfo.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/chromInfo.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gold.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gold.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gold.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gap.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gap.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/gap.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/oregannoAttr.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/oregannoAttr.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/oregannoAttr.frm /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/refLink.MYD /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/refLink.MYI /var/lib/mysql/hg19
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hg19/refLink.frm /var/lib/mysql/hg19
customTrash
hg19
hgFixed
hgcentral
proteinDB
proteins
proteome
swissProt
uniProt
#因此分别利用以下语句下载这些数据库
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/hgFixed/ /var/lib/mysql/hgFixed
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/uniProt/ /var/lib/mysql/uniProt/sp120323/
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/swissProt/ /var/lib/mysql/uniProt/sp111004/
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/proteome/ /var/lib/mysql/proteinDB/proteins120806/
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/mysql/proteins/ /var/lib/mysql/proteinDB/proteins111004/
sudo mkdir /gbdb
chmod 755 -R /gbdb
sudo rsync -avzP rsync://hgdownload.cse.ucsc.edu/gbdb/hg19/ /gbdb/hg19/
1、切记不要在虚拟机下的linux(无论是Redhat还是Ubuntu)系统中本地化UCSC，会遇到mysql占用大量CPU的问题。2、UCSC会遇到一个问题：Can’t connect to local MySQL server through socket ‘/var/lib/mysql/mysql.sock’ (13)。
可能是由于mysql版本的更新，.sock文件放的位置不同了。解决方法：
sudo ln -s /var/run/mysqld/mysqld.sock /var/lib/mysql/mysql.sock
sudo chmod 666 /var/lib/mysql/mysql.sock
sudo chmod 755 /var/lib/mysql/
Warning/Error(s):
Couldn't connect to database hg19 on localhost as user.
Access denied for user 'user'@'localhost' to database 'hg19'
说明目前还没有hg19数据库，接下来构建hg19并下载必要的文件，有时提示hg19.*，这说明缺少某个表，可以单独下载此表。
3、如果出现Warning/Error(s):External file /gbdb/hg19/multiz100way/maf/chr21.maf cannot be opened or has wrong size. Old size 8328154035, new size -1, error No such file or directory
则说明gbdb文件夹下缺少相应的文件，下载即可。
4、如果已经下载了本地化相关的大量文件，完全可以通过复制的办法来重新本地化，无需再次从UCSC网站下载。
5、如果本地根目录空间太小，无法存放mysql和gbdb两个超级文件夹，可以通过软链接（ln）的方式将其转移至较大的空间内。
（1）转移mysql目录：
cp -aR /var/lib/mysql/. /pub3/mysql
cd /var/lib/
mv mysql mysql_bei #将原有的mysql重命名，作为备份以防万一，同时也为成功做软链接做准备
ln -s /pub3/mysql /var/lib/mysql
chown -R mysql:mysql /pub3/mysql
service mysql start
cp -aR /gbdb/. /pub3/gbdb
mv gbdb gbdb_bei #以防万一
ln -s /pub3/gbdb /gbdb
其他供参考的UCSC本地化教程：
UCSC Genome Browser 本地化官方指南 （详细wiki）
UCSC Genome Browser 本地化Google论坛
UCSC Genome Browser 本地安装指南 by 卢金龙
UCSC Genome Browser 本地化安装指南（Ubuntu）by 吕杰
For any more questions, please contact hongbo919@gmail.com.
https://blog.sciencenet.cn/blog-97949-840924.html
下一篇：从一则招聘启事看生物信息研究者应掌握的十种能力
