piwi RNA / good practices

[raphael123]

Hi ! I had hard time finding a good piwi database so here is what I did, you can use it and tell me if it seems reasonable.
http://pirnabank.ibab.ac.in is the most used database, thez give you a list of fasta file corresponding to HG18 coordinates so I made this script to build a gff from their website:

for i in X Y $(seq 1 21); do 
  wget http://pirnabank.ibab.ac.in/cgi-bin/chr_result.pl?org=Homo_sapiens\&table=Human37_piRNA\&chr=$i\&from=0\&to=1000000000;
  mv chr_result.pl?org=Homo_sapiens\&table=Human37_piRNA\&chr=$i\&from=0\&to=1000000000 $i.html;
done 2> /dev/null >/dev/null


(echo "## gff-version 3"
echo "## date: $(date +%Y_%m_%d)"
echo "## Chromosomal coordinates of piwi RNA according to piRNAbank catin hsa genome"
for i in $(seq 1 21) X Y; do
cat $i.html  | sed ':a;N;$!ba;s/\n/ /g' | sed 's/table/\n/g' |  grep pir_info | tr ':\-=><|' ' '  | sed 's/  */ /g'  > temp 
grep -v Accession temp  | awk '{print $19,"NOID",$73,$74,$75,$76}'> temptemp
grep Accession temp | cut -d' ' -f20,61,93,94,95,96 >> temptemp
cat temptemp   | tr -d '[]"' | sort -nk3 > temp
cat temp| awk '{strand="-";if($6=="Plus"){strand="+"};printf "chr%s\tpiRNAbank\tpiwi\t%i\t%i\t.\t%s\t.\tID=%s;Name=%s\n",$3,$4,$5,strand,$2,$1}'
done
)  > pirna.gff