Closed mult1fractal closed 4 years ago
Hab den Docker für PhySpi fertig:
FROM continuumio/miniconda2
RUN apt-get update && apt install -y procps git wget make gcc g++ unzip && apt-get clean && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
RUN conda config --add channels conda-forge && \
conda config --add channels bioconda && \
conda config --add channels default
# Package randomForest in R - version 4.5-36 or later
RUN conda install -c conda-forge r-randomforest && \
conda install -c conda-forge/label/gcc7 r-randomforest && \
conda install -c conda-forge/label/cf201901 r-randomforest && \
conda install -c scikit-learn
#get git
RUN git clone https://github.com/linsalrob/PhiSpy.git
RUN cd /PhiSpy && \
make
RUN cd /PhiSpy && \
chmod a+x *
ENV PATH /PhiSpy:$PATH
(base) root@3629ca1b762e:/PhiSpy/Test_Organism/160490.1# ls Features TAXONOMY annotations contigs GENOME TAXONOMY_ID assigned_functions
(base) root@f7b9fccd7393:/# PhiSpy.py -i /home/mike/Documents/De-Phage/local_Docker_dev/testfolder/all_drafts/OX2_draft.fa -o outputdir -t 25 Cannot open /home/mike/Documents/De-Phage/local_Docker_dev/testfolder/all_drafts/OX2_draft.fa/contigs
funktioniert nicht
/foobar/OX2_Draft.fa existiert aber die msg zeigt das da noch /contigs irgenwo sein soll... gibts aber nicht
muss wahrscheinlich unsere .fa vorher noch mit SEED /RAST annotieren
% ./PhiSpy.py -i organism_directory -o output_directory -c
where: 'output directory': Output directory is the directory where the final output file will be created.
'organism directory': The seed annotation directory for the input bacterial organism whose prophage(s) need to be identified.
You can download the SEED genomes from the PhAnToMe database
Or, If you have new genome, you can annotate it using the RAST server. After annotation, you can download the genome directory from the server.
Or, If you have the GenBank file (containing sequence) of the genome, you can convert it using the following command: % python genbank_to_seed.py GenBank_file.gb organism_directory
Now to run PhiSpy, use organism_directory as 'organism directory'.
die anwendung sieht complexer aus mit diesen seed genomen: http://www.phantome.org/Downloads/genomes/seed/ naja adde das dockerfile und bau den auf docker hub.
Versuche mal rauszubekommen was du alles machen muss um das analysieren. genbank files kann man mit prokka erstellen. das kann ich einbauen.
Please, create config file using phigaro-setup script
"1\n1\nn" && phigaro-setup --no-updatedb
lösenFrom continuumio/miniconda3
ARG DEBIAN_FRONTEND=noninteractive
RUN apt-get update && apt install -y procps git locate wget make gcc && apt-get clean && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
RUN conda config --add channels conda-forge && \
conda config --add channels bioconda && \
conda config --add channels default
####Dockerfile
Run conda install hmmer prodigal
#get git
Run git clone https://github.com/bobeobibo/phigaro.git
#install Phigaro
# printf "1\n1\nn" | phi..... beantwortet alle fragen
RUN cd phigaro && \
pip install phigaro && \
chmod a+x * && \
printf "1\n1\nn" | phigaro-setup --no-updatedb
# phigaro-setup --no-updatedb
# datenbank download dauert 30 min
ENV PATH /Phigaro:$PATH
#
# phigaro -f /phigaro/test_data/Bacillus_anthracis_str_ames.fna -e txt html stdout -o result
#ergebnisdatei ist result
build und Error-msg
Please select appropriate Prodigal location [1] /opt/conda/bin/prodigal [2] Add another path manually Choose your option (Enter for /opt/conda/bin/prodigal): Please select appropriate HMMER location [1] /opt/conda/bin/hmmsearch [2] Add another path manually Choose your option (Enter for /opt/conda/bin/hmmsearch):
@Stormrider935
Please select appropriate Prodigal location [1] /opt/conda/bin/prodigal [2] Add another path manually Choose your option (Enter for /opt/conda/bin/prodigal): Please select appropriate HMMER location [1] /opt/conda/bin/hmmsearch [2] Add another path manually Choose your option (Enter for /opt/conda/bin/hmmsearch):
printf '1\n1\n1\nn'
Please select appropriate Prodigal location [1] /opt/conda/bin/prodigal [2] Add another path manually Choose your option (Enter for /opt/conda/bin/prodigal): Please select appropriate HMMER location [1] /opt/conda/bin/hmmsearch [2] Add another path manually Choose your option (Enter for /opt/conda/bin/hmmsearch):
Problem solved.. maybe...
I tried the following base image: From continuumio/miniconda2
and not miniconda 3
build process seems to work | sitll downloading database
but it's python 2.7....
works finally with the provided Testdata... only in /Phigaro
folder
Path is not the Problem
WORKDIR is /Phigaro
, but not sure how nextflow will handle it
we cant go into the /phigaro workdir. we have a separate process dir in nextflow
"""
printf 'Y' | phigaro -f test_data/Bacillus_anthracis_str_ames.fna -o test_data/Bacillus_anthracis_str_ames.phg -p --not-open -c /phigaro/config.yml
"""
General thread to talk and solve Dockerfile issues