===================================================== INSTALLATION OF EXTERNAL SOFTWARE REQUIRED BY EST2uni ===================================================== In this section you can find instructions about the installation of the software required by EST2uni in most Linux distributions. Please note that some analysis tools are optional, provided that you don't want to run that analysis. We have tested the installation of all this software in Fedora Core 6, SuSE 8.2, Ubuntu 6.06, and Debian Sarge. We assume that you want to install EST2uni with the same user that will run it, and that you already know how to install standard software packages (libraries, modules, etc.) in Linux. ---------------------------------------------------------------------- Perl ---------------------------------------------------------------------- Perl is usually installed by default in every Linux distribution. You can check it by typing: > which perl If no output is displayed, Perl is not installed. Please refer to your Linux distribution for specific installation instructions. The Perl standard database interface module perl-DBI (http://dbi.perl.org) must also be installed. It comes with every Linux distribution, so refer to your Linux distribution for specific installation instructions. ---------------------------------------------------------------------- Apache ---------------------------------------------------------------------- The web server Apache comes with every Linux distribution, so please refer to your distribution for installation instructions. ---------------------------------------------------------------------- MySQL ---------------------------------------------------------------------- MySQL is a popular relational database management system (RDMS) which comes with every Linux distribution. Please refer to your distribution for installation instructions. ---------------------------------------------------------------------- PHP ---------------------------------------------------------------------- PHP comes with every Linux distribution. Please refer to your distribution for installation instructions. The PHP module for Apache also needs to be installed. It comes with every Linux distribution, so please refer to your distribution for installation instructions. Also you need to install the PHP interface to MySQL to allow PHP to query the data stored in the MySQL database. The php-gd module is optional, but if you install it some nice charts will be displayed in your web site. ---------------------------------------------------------------------- Bioperl ---------------------------------------------------------------------- Bioperl can easily be installed from the CPAN Perl repositories, following the Bioperl site indications: > perl -MCPAN -e shell cpan> d /bioperl/ cpan> install B/BI/BIRNEY/bioperl-1.4.tar.gz If some tests fail, force the installation: cpan> force install B/BI/BIRNEY/bioperl-1.4.tar.gz Ubuntu and Debian have bioperl in their repositories and installation is much easier. ----------------------------------------------------------------------- phred ----------------------------------------------------------------------- 1.- Ask for phred distribution file to its authors (http://www.phrap.org/consed/consed.html#howToGet). 2.- Create a new directory: > mkdir phred 3.- Uncompress the tar file: > tar -xvzf phred-dist-020425.c-acd.tar.Z -C phred 4.- Compile the software: > cd phred > make 5.- Copy the executable to somewhere in the path: > cp phred /usr/local/bin/ 6.- Change the permissions: > chmod 755 /usr/local/bin/phred 7.- Copy the file phredpar.dat to a local configuration directory (e.g., /usr/local/etc): > cp phredpar.dat /usr/local/etc/ 8.- Add the PHRED_PARAMETER_FILE environment variable pointing to phredpar.dat file to your .bashrc file: > echo 'export PHRED_PARAMETER_FILE=/usr/local/etc/phredpar.dat' >> ~/.bashrc 9.- Check the installation by asking for the documentation: > phred -h 10.- When analyzing chromatograms, phred could give an error similar to: "unknown chemistry (DT3700POP5{BDv3}v1.mob) in chromat /home/ubuntu/test/libraries/sgn/TBFJQ34TH" In that case you should add a line with that chemistry to the phredpar.dat file. ---------------------------------------------------------------------- Lucy ---------------------------------------------------------------------- 1.- Download lucy from ftp://ftp.tigr.org/pub/software/Lucy/: > wget ftp://ftp.tigr.org/pub/software/Lucy/lucy-1.19p.tar.gz 2.- Uncompress the distribution: > tar -xvzf lucy-1.19p.tar.gz 3.- Compile the source code; > cd lucy-1.19p > make 4.- Copy the executable to your path: > cp lucy /usr/local/bin/ 5.- Check the executable: > lucy ---------------------------------------------------------------------- RepeatMasker ---------------------------------------------------------------------- 1.- Download the software from http://www.repeatmasker.org/RMDownload.html: 2.- Unpack: > tar -xvzf RepeatMasker-open-3-1-6.tar.gz 3.- Move the scripts to somewhere in your path: > mv RepeatMasker /usr/local/ 4.- You need Repbase from http://www.girinst.org/. Once that you register on its web page you will be allowed to download the repbase. 5.- Unpack the repbase distribution for RepeatMasker: > tar -xvzf repeatmaskerlibraries-YYYYMMDD.tar.gz 6.- Copy the library file to the Libraries directory in RepeatMasker installation location: > cp repeatmaskerlibraries-YYYYMMDD/RepeatMaskerLib.embl /where_you_installed_RepeatMasker/Libraries/ 7.- RepeatMasker needs cross_match or WU-BLAST to run: 7a.- cross_match: - cross_match is distributed with phrap; you can ask for it at http://www.phrap.org/consed/consed.html#howToGet - Copy the cross_match binary to somewhere in your path cp cross_match /usr/local/bin/ 7b.- WU-BLAST: - Ask for the licence at their web page (http://blast.wustl.edu/). - Uncompress the file: > mkdir wublast > tar -xvzf blast2.linux26-i686.tar.gz -C wublast > mv wublast /usr/local/ 8.- Run the configure script for RepeatMasker: > cd /usr/local/RepeatMasker > perl ./configure 11.- Add RepeatMasker to your path: > echo "export PATH=/usr/local/RepeatMasker:$PATH" >> ~/.bashrc 12.- Check that RepeatMasker is working: > RepeatMasker ---------------------------------------------------------------------- SeqClean ---------------------------------------------------------------------- 1.- Get SeqClean from http://compbio.dfci.harvard.edu/tgi/software/ 2.- Unpack the distribution: > tar -xvzf seqclean.tar.gz 3.- Move the SeqClean directory to /usr/local: >mv seqclean /usr/local/ 4.- Add the SeqClean directory to your path: > echo "export PATH=$PATH:/usr/local/seqclean:/usr/local/seqclean/bin" >> ~/.bashrc 5.- Fix the seqclean.psx permissions at the bin subdirectory: > chmod a+x seqclean.psx ---------------------------------------------------------------------- NCBI Blast ---------------------------------------------------------------------- 1.- Download the BLAST executable for your architecture from the NCBI blast download site (ftp://ftp.ncbi.nih.gov/blast/executables). 2.- Unpack the distribution: > tar -xvzf blast--.tar.gz 3.- Move the files to a proper location: > mv blast- /usr/local/blast 4.- Add the bin directory to the path: > echo "export PATH=$PATH:/usr/local/blast/bin" >> ~/.bashrc 5.- Check that everything is OK: > blastall 6.- Be sure to set the BLASTDB environment variable pointing to the directory where the BLAST formatted databases should be placed: > echo "export BLASTDB=/usr/local/blast_dbs" >> ~/.bashrc Once again, Ubuntu and Debian offer an easier way to install BLAST: > apt-get install blast2 ---------------------------------------------------------------------- UniVec ---------------------------------------------------------------------- 1.- Get UniVec database from NCBI: > wget ftp://ftp.ncbi.nih.gov/pub/UniVec/UniVec_Core 2.- Move UniVec to the BLAST databases location: > mv UniVec_Core /usr/local/blast_dbs/ 3.- UniVec should be formatted for BLAST: > formatdb -i UniVec_Core -V -p F -o (NOTE: NCBI's BLAST must be installed to be able to run 'formatdb') 4.- Check that the operation was successful: > cat formatdb.log ---------------------------------------------------------------------- TGICL ---------------------------------------------------------------------- 1.- Get TGICL from TIGR at http://compbio.dfci.harvard.edu/tgi/software/: > wget ftp://occams.dfci.harvard.edu/pub/bio/tgi/software/tgicl/tgicl_linux.tar.gz 2.- Unpack: > tar -xvzf tgicl_linux.tar.gz 3.- Copy the files to a proper location: > cp -r tgicl_linux /usr/local/tgicl 4.- Add the directory and the bin subdirectory to the your path: > echo "export PATH=$PATH:/usr/local/tgicl:/usr/local/tgicl/bin" >> ~/.bashrc 5.- Check that tgicl and cdbyank is working: > tgicl > cdbyank ---------------------------------------------------------------------- ESTScan ---------------------------------------------------------------------- 1.- Download ESTScan: > wget http://www.isrec.isb-sib.ch/ftp-server/ESTScan/ESTScan-2.0b.tar.gz 2.- Unpack the distribution: > tar -xvzf ESTScan-2.0b.tar.gz > cd BTLib-2.0b/ESTScan 3.- Open the ESTScan script and change the 307 line: from: $pSeq =~ s/[acgt]//g; # Remove lowercases... to: $pSeq =~ s/[acgtn]//g; # Remove lowercases... 4.- Install the BTLib distribution: > cd .. > perl Makefile.PL > make > make install 5.- Make sure that path to the perl interpreter is right on the first line of the ESTScan script. In our case is /usr/bin/perl. 6.- Move ESTScan to a proper location: > mv ESTScan /usr/local/ESTScan 7.- Add it to your path. 8.- Check that everything is correct: > ESTScan ---------------------------------------------------------------------- sputnik-like ---------------------------------------------------------------------- The original sputnik can be found at http://espressosoftware.com/pages/sputnik.jsp. This is not the version used by EST2uni. We provide a modified version that you can found in external_software/sputnik. 1.- Compile it. > cc -o sputnik sputnik.c 2.- Put it in your path: > mv sputnik /usr/local/bin 3.- Check that everything is OK: > sputnik ---------------------------------------------------------------------- go-perl ---------------------------------------------------------------------- For installation using CPAN: 1.- Start the CPAN installer > perl -MCPAN -e shell 2.- Install it: cpan> d /go-perl/ cpan> install C/CM/CMUNGALL/go-perl-0.05.tar.gz For manual installation, full instructions are given in http://www.godatabase.org/dev/go-perl/doc/go-perl-doc.html. ---------------------------------------------------------------------- ipcress ---------------------------------------------------------------------- 1.- Download 'exonerate' from http://www.ebi.ac.uk/~guy/exonerate/. Binaries are available for certain architectures. Here we give the instructions for the installation from source. 2.- Extract it: > tar -xvzf exonerate-1.2.0.tar.gz 3.- Be sure that 'glib' library is installed (it is required for next step): 4.- Compile it: > ./configure > make > make install 5.- Check it: > ipcress ---------------------------------------------------------------------- HMMER ---------------------------------------------------------------------- 1.- Download the most suitable HMMER distribution from http://hmmer.janelia.org/wiki/WikiStart. 2.- Read the nice HMMER documentation. 3.- Unpack the distribution: > tar -xvzf hmmer-2.3.2.bin.intel-linux.tar.gz 4.- Install it: > cd hmmer-2.3.2.bin.intel-linux/ > ./configure > make install 5.- Set the HMMERDB environment variable to the directory where the HMMER ready databases will be stored. > echo "export HMMERDB=/usr/local/hmmer_dbs" >> ~/.bashrc 5.- Check that everything went OK: > hmmpfam HMMER uses hidden Markov models to recover functional domains. So you need to install one of such databases. The analysis could work with Pfam. From the pfam FAQ: "You will need a copy of at least the file Pfam_ls from the Pfam FTP site, and possibly also the Pfam_fs file as well. The Pfam_ls file contains all the Pfam models for finding global or complete matches to the domain or family. Whereas the Pfam_fs file contains models for finding fragmentary matches to domains and families." 6.- Download the file Pfam_ls.gz. 7.- Unpack it and put it in the HMMERDB directory. > gunzip Pfam_ls.gz 8.- You can rename it to a more convenient name. > mv Pfam_ls pfam ----------------------------------------------------------------------