This page was last updated on Monday, 10-Apr-2017 09:19:59 CDT

Installation notes for PlantTFcat version April 30, 2014

Home Page

Prerequisites

java

Expat version 1.x, specifically we need libexpat.so.0

1. $ wget -N http://downloads.sourceforge.net/project/expat/expat/1.95.8/expat-1.95.8.tar.gz
2. $ tar -zxvf expat-1.95.8.tar.gz
3. $ cd expat-1.95.8
4. $ ./configure
5. $ make
6. $ sudo make install
   …
   Libraries have been installed in:
   /usr/local/lib
   …
7. Cleanup
   $ cd .. ; rm expat-1.95.8 -rf

Installation

1. Note, no apparent version number, file date is April 30, 2014
2. $ cd /programinstallers/
3. $ wget -N http://plantgrn.noble.org/PlantTFcat/download/PlantTFcat_pipeline.tar.gz
4. $ cd /usr/local/bin
5. $ tar -zxvf /programinstallers/PlantTFcat_pipeline.tar.gz
6. Some directory permissions need to be altered to allow users to make temporary files when running analyses
   $ sudo chmod -R 777 /usr/local/bin/PlantTFcat/session /usr/local/bin/PlantTFcat/miniIPR/tmp
7. Not added to default PATH, because instructions specify (in bold) to always use the full path
8. Test
   $ /usr/local/bin/PlantTFcat/PlantTFcat.sh

   usage: ./PlantTFcat.groovy -q <Query Sequence File> [--nodefile=]
                              [--minslicesize=] [--deltmp]
                              [--sessionid=1234567]
    -b,--base <arg>        Base path of pipeline
    -deltmp                Delete user-submitted tmp file from  web server.
    -E,--expect <arg>      Expect value of complementarity, default: 0.01
    -email <arg>           We will remind you by email upon completion
    -emailreceiver <arg>   Reply email address
    -emailsender <arg>     Sender email address
    -h,--help              print help information
    -hostname <arg>        hostname of web server
    -minslicesize <arg>    Minimum slice size of Biogrid, default will be
                           automatically determined by BioGrid
    -nodefile <arg>        Node configration file for BioGrid, pipeline will
                           run multi-threads within single computer without
                           it
    -port <arg>            port of web server
    -q,--query <arg>       Query sequence file full path
    -remoteaddr <arg>      remote user IP address, just for recording,
                           optional
    -sessionid <arg>       unique session ID, can be specified by user
    -smtp <arg>            smtp address

9. To resolve error groovy: JAVA_HOME not set and cannot find javac to deduce location, please set JAVA_HOME.
   $ which java
   /usr/bin/java
10. Note that /bin/java part is not included
    $ sudo nano /etc/profile
    …
    # for PlantTFcat
    export JAVA_HOME=/usr Don't leave this defined, breaks other programs!
    …
11. More errors encountered

     Caught: java.lang.OutOfMemoryError: unable to create new native thread
            at PlantTFcat.run(PlantTFcat.groovy:122)
    Exception in thread "Thread-54" java.lang.OutOfMemoryError: unable to create new native thread

    Fatal exception (source file /homes/hunter/software/source/hmmer/hmmer-3.0b3/easel/esl_threads.c, line 129):
    thread creation failed
    FATAL: Failed to create thread 19; return code 11

    I might have resolved both of these errors by increasing the java memory from 2048m to 6g in PlantTFcat.sh, increasing the maximum number of processes with ulimit -u 4096 (it was 1024 before), and increasing the maximum number of open file descriptors in /etc/security/limits.conf, before it was 10000 and I increased to 40000. But I am not sure which (if any) of these fixed it.