Some colleagues have been asking about installing and using the fabulous Prokka software on a compute cluster. Without root privledges this can be somewhat tricky if you don’t know where to start. Particularly, on our compute cluster the default Perl version is old, and you don’t have the option to install custom modules due to permissions restrictions. Here’s how you get a working standalone Perl environment that you can update and modify freely and how you install the Prokka perl module dependencies. We’ll be using the glorious Perlbrew.

Setting up PERL5 environment

cd # change to your home directory
wget -O - | bash
perlbrew install 5.24.0
cpan App::cpanminus

Installing Prokka dependencies

These are the required perl modules for Prokka

cpanm Time::Piece XML::Simple Bio::Perl Digest::MD5

If you are interested in prediciting rRNAs then I would recommend installing the barrnap program

tar zxvf barrnap-0.7.tar.gz

Finally, add barrnap-0.7/bin to your PATH

Installing Prokka itself

git clone # clone the prokka git repository into current directory

Add prokka-1.1X/bin to your PATH

prokka --setupdb

Good to go!