#!/usr/bin/env bash

if [ -z "$foreground" ] && [ -z "$background" ]; then 
	if (( $# < 2 )); then  
		echo "please supply two genomes, foreground and background"
		exit 1
	fi;

	: ${foreground=$1}
	: ${background=$2}
fi


if [[ ! -f $foreground ]]; then
	echo "Could not open $foreground"
	exit 1
fi

if [[ ! -f $background ]]; then
	echo "Could not open $background"
	exit 1
fi

# output directory 
: ${output_directory=`basename $foreground`_`basename $background`}

# temp directory 
: ${tmp_directory=$output_directory/.tmp}

# directory to store our counts and sorted counts
: ${counts_directory=$tmp_directory}

# range of mers, min and max 
: ${min_mer_range=6}
: ${max_mer_range=12}

# max mer distance, the distance between two mers in our selected outputs
: ${max_mer_distance=5000}

# min/maximum kmer meling point
: ${max_melting_temp=30}
: ${min_melting_temp=0}

# minimum mer count
: ${min_mer_count=0}

# maximum mers to pick
: ${max_select=15}

# maximum mers to check
: ${max_check=35}

# mers to specifically IGNORE, space delimited
: ${ignore_mers=''}

# maximum number of mers that are consecutively binding
: ${max_consecutive_binding=4}

export ignore_mers
export min_mer_range
export max_mer_range

export max_select

export min_mer_count
export max_mer_distance

export max_melting_temp 
export min_melting_temp 


# Make our output directory
if [ ! -d $output_directory ]; then
	mkdir $output_directory
fi

# Make our counts directory
if [ ! -d $counts_directory ]; then
	mkdir $counts_directory
fi

# Make our temporary directory
if [ ! -d $tmp_directory ]; then
	mkdir $tmp_directory
fi


current_run=$output_`date +%s`
mkdir -p $output_directory/$current_run 

for fasta_file in $foreground $background; do

	counts=$counts_directory/$(basename $fasta_file)
	tmp=$tmp_directory/$(basename $fasta_file)

	echo pre-processing $fasta_file

	# check if our preprocessed file exists
	if [[ ! -f $tmp ]]; then
		echo "> pre processed $fasta_file" >> $tmp
		cat $fasta_file | grep -v "^>" | tr -d '\n' >> $tmp
	fi

	# run counts if they haven't been created 
	rm $counts-counts
	for mer in `seq $min_mer_range $max_mer_range`;	do 
		if [ ! -e $counts-counts-$mer ]; then
			echo checking $mer mers for $fasta_file
			kmer_total_count -i $tmp -k $mer -l -n >> $counts-counts-$mer
		else 
			echo "$mer mers already done for $fasta_file"
		fi
		
		cat $counts-counts-$mer >> $counts-counts
	
	done
done

fg_counts=$counts_directory/$(basename $foreground)-counts
bg_counts=$counts_directory/$(basename $background)-counts

fg_tmp=$tmp_directory/$(basename $foreground)
bg_tmp=$tmp_directory/$(basename $background)

selected=$output_directory/$current_run/selected-mers

# remove ignored mers
if [ "$ignore_mers" ]; then
	echo "removing ignored mers: " + $ignore_mers
	for mer in $ignore_mers; do
		sed -i '/^'$mer'\t/d' $fg_counts
		sed -i '/^'$mer'\t/d' $bg_counts
	done
fi

echo "outputing current run parameters"
for var in ignore_mers min_mer_range max_check cpus max_consecutive_binding max_mer_range max_select min_mer_count max_mer_distance max_melting_temp min_melting_temp foreground background; do 
 echo $var "${!var}" >> $output_directory/$current_run/parameters
done;

echo "checking if mers are within the melting range $min_melting_temp $max_melting_temp"
cat $fg_counts | filter_melting_range $min_melting_temp $max_melting_temp > $output_directory/$current_run/$foreground-counts-non-melting 
cat $bg_counts | filter_melting_range $min_melting_temp $max_melting_temp > $output_directory/$current_run/$background-counts-non-melting 

echo "filtering out elements that have more consecutive binding mers than allowed by default $max_consecutive_binding"
cat $output_directory/$current_run/$foreground-counts-non-melting | filter_max_consecutive_binding.py $max_consecutive_binding > $output_directory/$current_run/$foreground-counts-filtered-binding
cat $output_directory/$current_run/$background-counts-non-melting | filter_max_consecutive_binding.py $max_consecutive_binding > $output_directory/$current_run/$background-counts-filtered-binding

echo "scoring mer selectivity"
select_mers.py $output_directory/$current_run/$foreground-counts-filtered-binding $fg_tmp $output_directory/$current_run/$background-counts-filtered-binding $bg_tmp > $selected
	
echo "scoring top mers based on selectivity"
score_mers.py $selected $fg_tmp $bg_tmp $output_directory/$current_run/scores-output