How to use standalone PHACTS

[Emily Kaelin]

Hello,

I'd like to use standalone PHACTS to classify phages in my dataset (I have several hundred to classify). Could you please clarify how the different perl scripts are meant to be used? Specifically, I would like to make sure I'm running them in the correct order, with the correct input.

Thanks very much!

Emily Kaelin

[Kate]

Hi, the scripts are just there in case you wish to create your own training data. 

The phacts.pl script should work after uncompressing the tarball, except you will need to edit line 37 of the script to point to the location of fasta35 (or fasta36).

    my $fasta_path = "/SOMEPATH/fasta36/bin/fasta36";

If you don't have the fasta binaries they can easily be installed via the commands:

    $ git clone https://github.com/wrpearson/fasta36.git

    $ cd fasta36/src

    $ make -f ../make/Makefile.linux_sse2 all

And then to run it:

    $ ./phacts.pl --file genomes/101570.3.fas --classes classes_lifestyle
    Predictions for genomes/101570.3.fas
    Class             probability                       standard deviation
    -----                  -----------                             ------------------
    Temperate    0.674025974025974    0.0697077957422893
    Lytic               0.325974025974026    0.0697077957422893
 

[Kate]

Oh, you will also need R and the R library 'randomForest'

[Emily Kaelin]

Thanks for the explanation! That all makes sense.

Emily