create_relations_receptors_files

#This subroutine is called from main program Vini. It creates receptors and relations files#


ORGANISM=`cat $WORKDIR/ORGANISM`              #getting cancer type from the main script
c=.xml
sav=.sav

CANCER_TYPE=`cat $WORKDIR/cancer_type`              #getting cancer type from the main script
CANCER_PATHWAY=$ORGANISM$CANCER_TYPE
PATHWAY_FILE=$CANCER_PATHWAY$c
PWD=`pwd`


cp $vini_dir/database/KEGG_cancer_pathways/$CANCER_PATHWAY/$PATHWAY_FILE $WORKDIR



cd $WORKDIR

b=`basename $PATHWAY_FILE`

echo -n "Preparing initial receptor / relation files..."

grep -P '^(?=.*entry)(?=.*gene)' $PATHWAY_FILE | sed 's/type=\"gene\"//'  | sed 's/\"//;s/\"//;s/\"//;s/\"//'  | sed 's/<entry//' | sed 's/id=//' | sed 's/name=//' > $WORKDIR/receptors

grep -P '^(?=.*entry)(?=.*compound)' $PATHWAY_FILE | sed 's/type=\"compound\"//' | sed 's/\"//;s/\"//;s/\"//;s/\"//'  | sed 's/<entry//' | sed 's/id=//' | sed 's/name=//' | sed 's/cpd://;s/cpd://;s/cpd://;s/cpd://;s/cpd://;s/cpd://;s/cpd://'>> $WORKDIR/receptors

grep -P '^(?=.*entry)(?=.*relation)' $PATHWAY_FILE | sed 's/<relation entry1=//;s/entry2=//;s/type=//;s/>//;s/\"//;s/\"//;s/\"//;s/\"//;s/\"//;s/\"//' | sed 's/rel//'  > $WORKDIR/relations #Create relations file

rm -f $PATHWAY_FILE

#KEGG entry and relation types: https://www.genome.jp/kegg/xml/docs/
#PPrel 	protein-protein interaction, such as binding and modification
#GErel 	gene expression interaction, indicating relation of transcription factor and target gene product
#ECrel 	enzyme-enzyme relation, indicating two enzymes catalyzing successive reaction steps
#PCrel 	protein-compound interaction
#maplink 	link to another map

while read -r line_relations                                #we will keep only relations within given pathway
do
         source_ID=`echo $line_relations | awk '{print $1}'`        #get source ID from relation
         target_ID=`echo $line_relations | awk '{print $2}'`        #get target ID from relation

         rec1_ID=0                                                  #search for source ID in the receptor file
         while read -r line_receptors                               #read line from receptor file
         do
             receptor_ID=`echo $line_receptors | awk '{print $1}'`  #get source ID from line
             if  [[ $receptor_ID -eq source_ID ]]
             then
                 rec1_ID=$source_ID ; break
             fi
         done < $WORKDIR/receptors

         rec2_ID=0                                                  #search for target ID in the receptor file
         while read -r line_receptors                               #read line from receptor file
         do
             receptor_ID=`echo $line_receptors | awk '{print $1}'`  #get source ID from line
             if  [[ $receptor_ID -eq target_ID ]]
             then
                 rec2_ID=$target_ID ; break
             fi
         done < $WORKDIR/receptors

         if  [[ $source_ID -eq $rec1_ID ]] && [[ $target_ID -eq $rec2_ID ]]
         then
              echo $line_relations >> $WORKDIR/temp_buf                                #write line in temp_buf
         fi
         echo -n "."

done < $WORKDIR/relations

mv $WORKDIR/temp_buf $WORKDIR/relations
echo "done."

total=0 ; echo -n "Expanding receptor file..."
while read
do
for word in $REPLY
    do
        echo -n "."
        A[$total]=$word
        total=$(($total+1))
    done
done < $WORKDIR/receptors

rm -f $WORKDIR/receptors_expanded entry_id
for i in "${A[@]}"
do
scale=${i}
if ! [[ "$scale" =~ ^[0-9]+$ ]]
        then
             entry_ida=`cat entry_id`
             printf "%s" ${entry_ida} " " ${scale}  >> $WORKDIR/receptors_expanded
             printf '%s\n' >> $WORKDIR/receptors_expanded
        else
             printf "%s" ${scale} > entry_id
        fi
done

rm entry_id
echo "done."