Blastx xml output parser

split_xml_blast_output.awk

#!/usr/bin/awk -f

# Author: Laurent Manchon ([email protected])
# Split big blast output in xml format into severals files
# Type split_xml_blast without parameters to see usage.


BEGIN{

      {
 if (ARGC==3 && ARGV[1] !~ "^[a-zA-Z]+$")
  {
      # max is number of sequences per output file
      max = ARGV[1]+0
      ARGV[1]=""
      } else
            {
              assert_exit = 1
            usage()
      }
}
      cpt=nb=1
      suffix=".xml"
      end="</BlastOutput_iterations>\n</BlastOutput>"
      begin="<?xml version=\"1.0\"?>\n<!DOCTYPE BlastOutput PUBLIC \"-//NCBI//NCBI BlastOutput/EN\" \"http://www.ncbi.nlm.nih.gov/dtd/NCBI_BlastOutput.dtd\">\n"
      begin=begin "<BlastOutput>\n<BlastOutput_program>blastx</BlastOutput_program>\n<BlastOutput_version>blastx 2.2.18 [Mar-02-2008]</BlastOutput_version>\n"
      begin=begin "<BlastOutput_reference></BlastOutput_reference>\n<BlastOutput_db>/home/data/blastdb/nr</BlastOutput_db>\n<BlastOutput_query-ID>lcl|1_0</BlastOutput_query-ID>\n"
      begin=begin "<BlastOutput_query-def></BlastOutput_query-def>\n<BlastOutput_query-len></BlastOutput_query-len>\n<BlastOutput_param>\n<Parameters>\n<Parameters_matrix>BLOSUM62</Parameters_matrix>\n"
      begin=begin "<Parameters_expect>0.1</Parameters_expect>\n<Parameters_gap-open>11</Parameters_gap-open>\n<Parameters_gap-extend>1</Parameters_gap-extend>\n<Parameters_filter>F</Parameters_filter>\n"
      begin=begin "</Parameters>\n</BlastOutput_param>\n<BlastOutput_iterations>"
}

function usage()
{

print "###################################################################################"
print "#   split_xml_blast --  split big blast output in xml format into severals files. #"
print "#   Performed in Awk v3.1         A.V. Aho, P.J. Weinberger, and B.W. Kernighan   #"
print "#   OS supported: *nix, Windows9x/NT                                              #"
print "###################################################################################"
print "#    Author: Laurent Manchon                                                      #"
print "#    If you have comments or questions, send to the author at:                    #"
print "#    [email protected]                                                      #"
print "###################################################################################"
print "#                                                                                 #"
print "#     This program takes a file containing blast result in XML format and split   #"
print "#     it into severals small files, as: split_xml_blast <nb> <input_filename>     #"
print "#     with <nb>: Number of sequences per output file                              #"
print "#                                                                                 #"
print "###################################################################################"

    exit 1
}

/<Iteration>/{
split(FILENAME,prefix,".")
file=prefix[1] "_"
output_file=file nb suffix
i=1
if(cpt==1){print begin >> output_file}
print $0 >> output_file
next
}

i==1{print $0 >> output_file}

/<\/Iteration>/{
                cpt++
                if(cpt==max+1){
                     print end >> output_file
                     close (output_file)
                     nb++
                     cpt=1
                     i=0
                     next
                      }

}

END {
     if (assert_exit) exit 1
 print "\nYour input file",FILENAME,"has just been splitted into",nb,"files with",max,"sequences per file:\n"     
cmd="ls -1 "file"*.xml"
system(cmd)
close(cmd)
}