faircloth-lab
latest

Contents:

  • Lab Protocols
  • Computer Protocols
    • Sequencing
    • Assembly
    • Analysis
    • Snippets
    • Compilation/Installation
      • Compiling IQ-tree
      • Compiling Pargenes
      • Compiling RAxML-NG
      • Compiling Canu
      • Compiling Arks And Links
        • Steps
      • Compiling Supernova
    • Website
  • Data Protocols
  • Changelog
faircloth-lab
  • Docs »
  • Computer Protocols »
  • Compiling Arks And Links
  • Edit on GitHub

Compiling Arks And Links¶

Author:Brant C. Faircloth
Copyright:This documentation is available under a Creative Commons (CC-BY) license.

Steps¶

  1. Download and unpack arks

    wget -O arks-v1.0.4.tar.gz https://github.com/bcgsc/arks/archive/v1.0.4.tar.gz
    tar -xzvf arks-v1.0.4.tar.gz
    
  2. Download and compile a local copy of sparsehash

    # in $HOME/project/shared/src
    cd sparsehash/ && ./configure && make && cd ../
    
  3. Load the Boost and GCC 6 modules and set CC and CXX and CPPFLAGS

    module load boost/1.63.0/INTEL-18.0.0
    module load gcc/6.4.0
    
    export CC=`which gcc`
    export CXX=`which g++`
    export CPPFLAGS=-I$(readlink -f sparsehash/src)
    
  4. Enter the arks directory and run autogen.sh and configure and, finally, make install

    cd arks-1.0.4/
    ./autogen.sh
    ./configure --with-boost=/usr/local/packages/boost/1.63.0/INTEL-18.0.0/lib --prefix=$HOME/project/shared
    make install
    
  5. Apparently, one of the needed files does not get copies to bin, so:

    cd ~/project/shared/bin
    ln -s ../src/arks-1.0.4/Examples/makeTSVfile.py
    
  6. Download links and unzip it

    wget https://github.com/bcgsc/LINKS/releases/download/v1.8.7/links_v1-8-7.tar.gz
    tar -xvf links_v1-8-7.tar.gz
    
  7. We need to build the bloomfilter module for links. To compile with more modern versions of GCC (> v4), omit the -Dbool=char flag:

    module load perl/5.24.0/INTEL-18.0.0
    cd links_v1.8.7/lib
    rm -rf bloomfilter/
    git clone git://github.com/bcgsc/bloomfilter.git
    cd bloomfilter/swig
    swig -Wall -c++ -perl5 BloomFilter.i
    # omit the -Dbool=char flag in the following
    g++ -c BloomFilter_wrap.cxx -I/usr/local/packages/perl/5.24.0/INTEL-18.0.0/lib/5.24.0/x86_64-linux-thread-multi/CORE -fPIC -Dbool=char -O3
    g++ -Wall -shared BloomFilter_wrap.o -o BloomFilter.so -O3
    
    # test with
    perl test.pl
    
  8. Finally, create a conda environment that contains the remaning dependencies or arks:

    conda create -n scaffolding tigmint bwa samtools bedtools
    
Next Previous

© Copyright 2017-2019, Brant Faircloth and The Faircloth Lab Revision e4a8853e.

Built with Sphinx using a theme provided by Read the Docs.