Skip to content

Latest commit

 

History

History
147 lines (96 loc) · 7.12 KB

conda.rst

File metadata and controls

147 lines (96 loc) · 7.12 KB
Raw

Conda mediated Installation

I'd really like to build a bioconda installation package, but would need some help. You can however install quite a few of the dependencies with conda.

If you are on LINUX -- start here:

#If you do not have conda, install: download miniconda2 or miniconda3, miniconda3 shown
wget --quiet https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh
/bin/bash ~/miniconda.sh -b -p /conda/installation/path

#setup bioconda repository
conda config --add channels defaults
conda config --add channels etetoolkit
conda config --add channels bioconda
conda config --add channels conda-forge

#now create a conda environment and install dependencies
conda create -y -n funannotate python=2.7 numpy pandas scipy matplotlib seaborn \
    natsort scikit-learn psutil biopython requests blast rmblast goatools fisher  \
    bamtools augustus bedtools hmmer exonerate diamond>=0.9 tbl2asn ucsc-pslcdnafilter \
    samtools raxml trimal mafft>=7 iqtree kallisto>=0.46.0 bowtie2 infernal mummer minimap2 blat \
    trinity>=2.6.6 evidencemodeler pasa>=2.3 codingquarry stringtie gmap=2017.11.15 snap \
    ete3 salmon>=0.9 jellyfish>=2.2 htslib trnascan-se hisat2 glimmerhmm \
    trf perl-threaded perl-db-file perl-bioperl perl-dbd-mysql perl-dbd-sqlite \
    perl-text-soundex perl-scalar-util-numeric perl-data-dumper perl-dbi perl-clone \
    perl-json perl-logger-simple perl-hash-merge perl-yaml perl-pod-usage perl-getopt-long \
    perl-parallel-forkmanager perl-carp perl-soap-lite perl-class-inspector perl-app-cpanminus

#if you are going to use remote search also need LWP module (not on conda)
cpanm LWP

If you are on MacOS X -- start here:

#If you do not have conda, install: download miniconda2 or miniconda3, miniconda3 shown
wget --quiet https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh
/bin/bash ~/miniconda.sh -b -p /conda/installation/path

#setup bioconda repository
conda config --add channels defaults
conda config --add channels etetoolkit
conda config --add channels bioconda
conda config --add channels conda-forge

#now create a conda environment and install dependencies
conda create -y -n funannotate python=2.7 numpy pandas scipy matplotlib seaborn \
    natsort scikit-learn psutil biopython requests blast rmblast goatools fisher \
    bedtools hmmer exonerate diamond>=0.9 tbl2asn ucsc-pslcdnafilter \
    samtools raxml trimal mafft>=7 iqtree kallisto>=0.46.0 bowtie2 infernal mummer \
    evidencemodeler  gmap=2017.11.15 hisat2 blat minimap2 snap glimmerhmm  \
    ete3 salmon>=0.9 jellyfish>=2.2 htslib trnascan-se codingquarry \
    trf perl-threaded perl-db-file perl-bioperl perl-dbd-mysql perl-dbd-sqlite \
    perl-text-soundex perl-scalar-util-numeric perl-data-dumper perl-dbi perl-clone \
    perl-json perl-logger-simple perl-hash-merge perl-yaml perl-pod-usage perl-getopt-long \
    perl-parallel-forkmanager perl-carp perl-soap-lite perl-class-inspector perl-app-cpanminus

#if you are going to use remote search also need LWP module (not on conda)
cpanm LWP

MacOSX: Need to install bamtools/augustus/trinity/pasa manually:

Install bamtools/Augustus from here: https://github.com/nextgenusfs/augustus

Trinity: https://github.com/trinityrnaseq/trinityrnaseq

PASA: https://github.com/PASApipeline/PASApipeline

The above will automatically install most of the dependencies, below there are a few manual steps.

  1. Download/install GeneMark-ES/ET: (gmes_petap.pl must be in PATH) http://exon.gatech.edu/GeneMark/license_download.cgi
    • make sure to activate the license and move into proper location. you can test proper installation by running gmes_petap.pl in the terminal -- you should see help menu. Be careful of the shebang line, default is /usr/bin/perl which most likely is not what you want, more appropriate is /usr/bin/env perl
  2. Install RepeatMasker/RepeatModeler http://www.repeatmasker.org

2b. Download Repbase RepeatMasker Libraries if you have not done so already.

wget --user name --password pass http://www.girinst.org/server/RepBase/protected/repeatmaskerlibraries/RepBaseRepeatMaskerEdition-20170127.tar.gz
tar zxvf RepBaseRepeatMaskerEdition-20170127.tar.gz -C /path/to/repeatmasker/location
cd /path/to/repeatmasker/location
./configure

#Soft-link a repeatmasker utility script into the PATH (may not need to do this depending on install)
ln -s /path/to/repeatmasker/location/repeatmasker/util/rmOutToGFF3.pl /usr/local/bin/rmOutToGFF3.pl
  1. Setup Eggnog-mapper v4.5 or v5.0 [v5.0 is not being parsed properly yet in v1.5.3]
#clone the eggnog mapper repo into a location you have read/write access
git clone https://github.com/jhcepas/eggnog-mapper.git

#move into folder and setup - this will put into eggnog-mapper/data location
cd eggnog-mapper
download_eggnog_data.py

#finally add to your funannotate conda env so it is in path when env is activated
ln -s /path/to/eggnog-mapper/emapper.py /path/to/conda/envs/funannotate/bin/emapper.py

NOTE: MacOSX users -- the diamond version shipped with eggnog-mapper needs to be swapped out as the binary provided is compiled on linux. Run a small test with emapper.py to check functionality emapper.py -m diamond -i test.fa -o test

  1. Clone the funannotate repo and add to PATH
git clone https://github.com/nextgenusfs/funannotate.git

#add to PATH
ln -s /path/to/funannotate/funannotate /path/to/conda/envs/funannotate/bin/funannotate
  1. Run funannotate check --show-versions, fix any issues. You will need to export some ENV variables.
export EVM_HOME=/path/to/conda/envs/funannotate/opt/evidencemodeler-v1.1.1
export TRINITYHOME=/path/to/conda/envs/funannotate/opt/trinity-2.6.6
export PASAHOME=/path/to/conda/envs/funannotate/opt/pasa-2.3.3
export AUGUSTUS_CONFIG_PATH=/path/to/augustus/config
export GENEMARK_PATH=/path/to/gmes_petap_dir
export FUNANNOTATE_DB=/path/to/funannotateDB
  1. Setup funannotate databases, specify any location you have read/write access to to -d -- this is $FUNANNOTATE_DB
funannotate setup -d /path/to/DB
  1. If you want these ENV variables to be activated when you activate the conda environment, you can add them as a shell script to the the activate location of your environment, i.e. /path/to/conda/envs/funannotate/etc/conda/activate.d/ and then you can put the corresponding unset commands in the deactivate directory, i.e. /path/to/conda/envs/funannotate/etc/conda/deactivate.d/