Shell Programming and Scripting

WASH7P        17232,18267,18500,20564            17368,18362,18554,21139

chr1	14969	15038	Exon	WASH7P
chr1	17232	17368	Exon	WASH7P
chr1	17258	17368	Exon	WASH7P
chr1	17605	17742	Exon	WASH7P
chr1	18267	18362	Exon	WASH7P
chr1	18267	18366	Exon	WASH7P
chr1	18267	18369	Exon	WASH7P
chr1	18267	18379	Exon	WASH7P
chr1	18496	18554	Exon	WASH7P
chr1	18500	18554	Exon	WASH7P
chr1	18912	19139	Exon	WASH7P

 WASH7P        17232,18267,18500,20564            17368,18362,18554,21139  17232,18267,18500   17232,18267,18500

#! /bin/sh

awk '
# only process file1 and store all the mappings
FNR==NR {
        # store mapping of column 2 in file1
        n=split($2,a,",")
        for(i=1;i<=n;++i) {
                map2[$1,a[i]]=1
        }

        # store mapping of column 3 in file1
        n=split($3,a,",")
        for(i=1;i<=n;++i) {
                map3[$1,a[i]]=1
        }

        # store original file
        file1[$1]=$0

        next
}

# start to process file2
# if column2 in file2 exists in file1, concatenate the index to string str2
# if column3 in file2 exists in file1, concatenate the index to string str3
{
        if ( map2[$5,$2]==1 && match2[$5,$2]!=1 ) {
                str2[$5]=sprintf("%s,%s",str2[$5],$2)
                match2[$5,$2]=1
        }
        if ( map3[$5,$3]==1 && match3[$5,$3]!=1 ) {
                str3[$5]=sprintf("%s,%s",str3[$5],$3)
                match3[$5,$3]=1
        }
}
END {
        for ( i in file1 ) {
                print file1[i], substr(str2[i],2), substr(str3[i],2)
        }
}' file1 file2

awk 'NR==FNR{A[$2,$3]=1;next}
     { m=split($2,S,/,/);split($3,T,/,/);
       for(i=1;i<=m;i++) 
         if(A[S[i],T[i]]){
            $4=$4 ($4?",":x) S[i]
            $5=$5 ($5?",":x) T[i]
         }
     }1' file2 file1