ProDeGe: a computational protocol for fully automated decontamination of genomes

Question

[Mehmet]

Dear All:

I want to use the tool "ProDeGe: a computational protocol for fully automated decontamination of genomes". Any of you has used before?

http://jgi.doe.gov/automating-microbial-genome-sequence-decontamination/

I tried to find a manual, but I could not. I need an explanation about how to use.

Thank you.

[Brian Bushnell]

I'm familiar with it, but have not personally used it. Can you describe the project you are working on and the type of contamination you are experiencing?

[Mehmet]

hi; I am working on a worm project. I have its transcriptome data. I would like to check whether the data is contaminated by bacteria or not. I want to know how to do this by prodege. I need a step by step explanation.

Thank you.

[Brian Bushnell]

Everyone at JGI is on vacation because it's Thanksgiving, but on Monday I can talk to the people who use it routinely.

[GenoMax]

Why not use bbsplit with worm reference. That should leave all non-worm reads behind. Check the reads with blast afterwards.

[Mehmet]

Dear Brian:

I got the program from my supervisor. Can I ask you a few questions about the program?

[Brian Bushnell]

You can certainly ask, but like I said, I have never used it, I only work with people who use it. I know a little about how it works, though.

[Mehmet]

OK.

the program needs a configuration file, as it is suggested in README file.

I am pasting here: INPUT/OUTPUT ProDeGe takes as input a fasta file that includes contigs from contaminant origin and the taxonomy of the target organism, and creates a fasta file including only the contigs from the target organism. Several intermediary files are created during the process.

RUNNING prodege.sh <configuration file="">

EXAMPLE Two examples are included in the Examples folder. The config.cfg needs to be updated with the install location and the working directory. To run the example type: prodege.sh <config file="">

My questions are:

Should I create a configuration file in which I have to write bacteria names that I want to check in my transcriptome data?If yes, can I write general bacteria names?

I need help about how to run this analysis.

[Mehmet]

CONFIGURATION
The program takes as input a configuration file. Below are the fields that should be in the file:
TAXON_DISPLAY_NAME=         # This field is not required, put quotes around if spaces
TAXON_DOMAIN=               # Put quotes around taxon var assignments
TAXON_PHYLUM=               # because some have spaces in them
TAXON_CLASS=                # like "Bacteroidetes/Chlorobi group"
TAXON_ORDER=                # TAXON_X can be blank
TAXON_FAMILY=
TAXON_GENUS=            
INSTALL_LOCATION=           # Do not set this if you are using 'module load prodege'
WORKING_DIR=                # Prodege will create a folder $JOB_NAME in $WORKING_DIR
IN_FASTA=           # Full path of input fna
JOB_NAME=                   # Clean fasta file will be: ${WORKING_DIR}/${JOB_NAME}/${JOB_NAME}_output_clean.fna
RUN_GENECALL=<0 to skip, 1 to run>
RUN_BLAST=<0 to skip, 1 to run>
RUN_CLASSIFY=<0 to skip, 1 to run>
RUN_ACCURACY=<0 to skip, 1 to run>  #contigs must have "clean" or "contam" in their fasta's contig name
BLAST_THREADS=                      #Default is 8 
KMER_CUTOFF=                #Prodege is precalibrated with a cutoff, this field is optional

[Tonor]

According to their paper, the program should be able to download from here: https://prodege.jgi-psf.org/downloads/src

But the whole server seems to be down, I would suspect that is where the manual would be

[Mehmet]

I tried but I could not find anything about the program, including manual and program itself.

[Tonor]

Yeah the whole server seems to be down, the web version, the standalone version, the whole thing