Installing the workshop software
We are going to install the software needed for this workshop
- FLASH2 to overlap reads (https://github.com/dstreett/FLASH2)
- RDP to classify reads into taxon (https://github.com/rdpstaff/RDPTools)
- dbcAmplicons pipeline for processing amplicon sequences to abundance tables (https://github.com/msettles/dbcAmplicons)
optional
- bowtie2 (https://github.com/BenLangmead/bowtie2)
System Requirements
- git
- java jdk
- ant
- python virtualenv
- biom requires numpy
1. First, create a directory for the workshop:
cd
mkdir -p /share/workshop/$USER/mca_example
and two other directors
mkdir /share/workshop/$USER/mca_example/src
mkdir /share/workshop/$USER/mca_example/bin
2. Now lets add the new bin directory to our PATH in a dbcA_profile file
echo 'export PATH=/share/workshop/$USER/mca_example/bin:$PATH' > /share/workshop/$USER/mca_example/src/dbcA_profile
Then on the command line, execute the commands in the dbcA_profile file using source.
source /share/workshop/$USER/mca_example/src/dbcA_profile
3. Install FLASH2 into src and link the executable into bin
cd /share/workshop/$USER/mca_example/src
git clone https://github.com/dstreett/FLASH2.git
cd FLASH2
make
ln -s /share/workshop/$USER/mca_example/src/FLASH2/flash2 /share/workshop/$USER/mca_example/bin/.
# test installation, should see help documentation
flash2 -h
4.a Install apache ant, need for RDP
cd /share/workshop/$USER/mca_example/src
curl http://mirrors.ibiblio.org/apache/ant/binaries/apache-ant-1.10.5-bin.tar.gz > apache-ant-1.10.5-bin.tar.gz
tar xzvf apache-ant-1.10.5-bin.tar.gz
ln -s /share/workshop/$USER/mca_example/src/apache-ant-1.10.5/bin/ant /share/workshop/$USER/mca_example/bin/.
4.b Install the Ribosomal Database Project (RDP) into src
module load java/jdk1.8 # Need the java jdk1.8 loaded
cd /share/workshop/$USER/mca_example/src
git clone https://github.com/rdpstaff/RDPTools.git
cd RDPTools/
git submodule init
git submodule update
make
might have problems retrieving data.tgz file … if so, try again … during 9-5 EST(!) …
test installation, should see help documentation for classify
java -jar classifier.jar classify
this should give you a “Command Error” because you didn't specify output, but it should give you a list of options. feel free to move on if the ‘make’ fails due to data.tgz file, we can fix this later
4.c Add the location of classifier.jar as a variable in our dbcA_profile file,
echo 'module load java/jdk1.8' >> /share/workshop/$USER/mca_example/src/dbcA_profile
echo 'export RDP_PATH=/share/workshop/$USER/mca_example/src/RDPTools' >> /share/workshop/$USER/mca_example/src/dbcA_profile
Then on the command line, execute the commands in the dbcA_profile file using source.
source /share/workshop/$USER/mca_example/src/dbcA_profile
5.a Setup a conda python environment for dbcAmplicons, in the src directory.
module load anaconda2
conda create -p /share/workshop/$USER/mca_example/src/dbcA_virtualenv python=2.7 pip
5.b This lets you set the virtual environment to activate on login by adding it to our dbcA_profile file.
echo 'module load anaconda2' >> /share/workshop/$USER/mca_example/src/dbcA_profile
echo '. /software/anaconda2/4.5.12/lssc0-linux/etc/profile.d/conda.sh' >> /share/workshop/$USER/mca_example/src/dbcA_profile
echo 'conda activate /share/workshop/$USER/mca_example/src/dbcA_virtualenv' >> /share/workshop/$USER/mca_example/src/dbcA_profile
echo 'export PYTHON_EGG_CACHE=/share/workshop/$USER/mca_example/src' >> /share/workshop/$USER/mca_example/src/dbcA_profile
Then on the command line, execute the commands in the dbcA_profile file using source.
source /share/workshop/$USER/mca_example/src/dbcA_profile
You should now see the text “(dbcA_virtualenv)” at the beginning of your prompt.
6. Install dbcAmplicons
cd /share/workshop/$USER/mca_example/src
easy_install pip
pip install biom_format
git clone https://github.com/msettles/dbcAmplicons.git
cd /share/workshop/$USER/mca_example/src/dbcAmplicons/
python setup.py install
test installation, should see help documentation
dbcAmplicons -h
You should see the options / usage message
Optional. Test dbcAmplicons
You could also test the dbcAmplicons installation by running the script, test_dbAmplicons.sh, under the tests folder (in dbcAmplicions).
cd /share/workshop/$USER/mca_example/src/dbcAmplicons/tests/
./test_dbAmplicons.sh
The script should show some ERRORs / WARNINGs (intended as test), but otherwise give stats after a few minutes
Lets Review
We created a directory for the workshop, in that directory we created two folders src and bin. We’ve installed FLASH2, RDP and dbcAmplicons. We’ve placed the executable for flash in a bin folder and added the folder to our PATH. We created an environment variable for the RDP classifier. We’ve created a conda python environment and then installed the dbcAmplicons package using setup.py.
- You should have verified all the software works by viewing their help docs
- Verify that the flash executable is indeed in the bin folder
- We’ve modified your PATH and added 1 new environment variable, RDP_PATH, verify the PATH and the new env variable.
- We added a multiple lines to your .bash_profile. How many lines?
Now log out, log back in and verify that each application still works. Ex.
flash2 -h
To verify RDP and dbcAmplicons use
java -jar $RDP_PATH/classifier.jar classify
dbcAmplicons -h
You can check the current version of everything with
dbcVersionReport.sh
We usually save the output of this in the project file to remind ourselves which versions of software were run for that project.
If for some reason installation failed let me know and we’ll get things fixed
These lines should be in your dbcA_profile
At any time you can initiate the environment for dbcAmplicons using.
source /share/workshop/$USER/mca_example/src/dbcA_profile
and you can specifically deactivate the conda environment with
conda deactivate