testing recent shorah directly from github

shorah-tut.sh

mkdir -p ~/shorah-test
cd ~/shorah-test


#
# install bioconda
#

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

# -b for batch (no question asked)
bash Miniconda3-latest-Linux-x86_64.sh -b -p ~/shorah-test/miniconda3


. ~/shorah-test/miniconda3/bin/activate 

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge

conda update -n base conda

#
# create conda environment with all of shorah's dependencies
#

conda create -p ~/shorah-test/shorah_env python=3.7 htslib biopython numpy gsl zlib bzip2 boost readline perl automake=1.15.0 autoconf pkg-config
conda activate  ~/shorah-test/shorah_env

#
# install shorah
#
# (allocaholic is the latest branch that will eventually made into version 2.0)

git clone https://github.com/cbg-ethz/shorah.git -b allocaholic
cd ~/shorah-test/shorah

aclocal -I m4
autoconf
automake -a -c


mkdir -p build
cd build

../configure PKG_CONFIG_PATH=$HOME/shorah-test/shorah_env/lib/pkgconfig/ CFLAGS="-O3 -ffast-math -Wall -pedantic -g" CXXFLAGS="-O3 -ffast-math -Wall -pedantic -g -std=gnu++11"  LDFLAGS="-Wl,-O1 -Wl,--as-needed -Wl,-rpath=$HOME/shorah-test/shorah_env/lib/" --with-boost=$HOME/shorah-test/shorah_env/ --prefix=$HOME/shorah-test/ 

make clean
make all
make install


#
# use shorah
#
## input: cohort_consensus.fasta -- reference genome against which samples are aligned
## input: REF_aln.bam / .bam.bai -- aligned and indexed samples
#
## param: -x limit samples per windows (increase it if you use more coverage than 100'000)
## param: -S sigma value to use when calling SNVs
## param: -w size of window (needs to be smaller than samples !)
#
## other userful:
## -r chrm:start-stop, --region chrm:start-stop
##                      region in format 'chr:start-stop', e.g.
##                      'chrm:1000-3000'
## -R {nnn} : fixed seed for reproducible runs.
## -c {nnn} : minimal coverage (ignore windows with less than nnn coverage)
## -s {nnn} : shift between windows
#

cd data

PATH=$HOME/shorah-test/bin:$PATH PYTHONPATH=$HOME/shorah-test/lib/python3.7/site-packages ~/shorah-test/bin/shorah shotgun --bam REF_aln.bam --fasta cohort_consensus.fasta -x 100000 -S 42 -w 102

#
# for visualisation
#

# support/
#
# contains, per window, the found haplotypes 
# posterior: bayesian posterior porbably (i.e.: after the run)
# ave_read: number of reads that support bases of this haplotype in average (average per base, because some samples only fill the window partially, hence the non-int number)

#
# for down stream further analysis
#
# .cor.fas
#
# the alignement sample from the BAM file, but correct based on corrected output (correct/) from each overlapping window