Parsing a file

cowboyrocks has asked for the wisdom of the Perl Monks concerning the following question:

Hi Monks
I am trying to catch the subsequent last line "Sbjct:". Suppose "Sbjct: 38348698" and "Sbjct: 94217074" lines in this example but
unable to do so. How can I catch these lines ?
My code looks like this :-


     if($_ =~ /Sbjct:\s*(\d+)\D+(\d+)*$/ .. /\n\n\n\AQuery\=/) 
                   { 
                         print "$_\n"
                   }
[download]

My input file looks like this:-

Query= 30 3'_UTR BP;A;C;G;T;OTHER
AGGCATAAACCACATCCAGCCACCTCCTTCTGATCAGCAGCAAAGCTGACGTTTTGATCTCCATCTGTCT
GATTCTTGTGTCTACTTCTCAGTTTACAACTCCAGTGGGAAAGAAAGAGCTTTATTTACAGACCCATAAA
AATCCCATCAGTGTCGTCCCCTGCTGAGAGGCCATGTGAGACCATATGGAAAAACAACAGCCATAATGGC
AGCATGGCAGTGGAAGGGTTTGTCTTGTGCCCAGGCCTTGCGGTCATGCAAGTTTCTTGTGGATCCTGTT

          (633 letters)



                                                                   Sco
+re     E
Sequences producing significant alignments:                        (Bi
+ts)  Value

gi|51511750|ref|NC_000021.7|NC_000021 Homo sapiens chromosome ...  124
+3  0E0

>gi|51511750|ref|NC_000021.7|NC_000021 Homo sapiens chromosome 21, ref
+erence assembly, complete sequence
          Length = 46944323

 Score = 1243 bits (629),  Expect = 0E0
 Identities = 632/633 (99%)
 Strand = Plus / Minus

Query: 1        AGGCATAAACCACATCCAGCCACCTCCTTCTGATCAGCAGCAAAGCTGACGTTT
+TGATCT 60
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 38348818 AGGCATAAACCACATCCAGCCACCTCCTTCTGATCAGCAGCAAAGCTGACGTTT
+TGATCT 38348759

Query: 61       CCATCTGTCTGATTCTTGTGTCTACTTCTCAGTTTACAACTCCAGTGGGAAAGA
+AAGAGC 120
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 38348758 CCATCTGTCTGATTCTTGTGTCTACTTCTCAGTTTACAACTCCAGTGGGAAAGA
+AAGAGC 38348699

Query: 121      TTTATTTACAGACCCATAAAAATCCCATCAGTGTCGTCCCCTGCTGAGAGGCCA
+TGTGAG 180
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 38348698 TTTATTTACAGACCCATAAAAATCCCATCAGTGTCGTCCCCTGCTGAGAGGCCA
+TGTGAG 38348639

Query= 72 3'_UTR BP;A;C;G;T;OTHER
CAAGAAACTATATAGGTATACACTTACGACTTCACAAAACCTATACTTAATATAGTAAATCTAAGTAAAC
ATGTATTACTCAAAGTAATATATTTAGAATTATGTATTAGTATAAGATCAGAATTGAATTTAAGTTGTTG
GTGACATCTGCATCATTTCATAGGATTAGAACTTACTCAAAATAATGTAAATCTTTAAAAATATAAATTA
GAATGACAAGTGGGAATCATAAATTAAACGTTAATGGTTTCTTATGCTCTTTTTAAATATAGAAATATCA
          (897 letters)



                                                                   Sco
+re     E
Sequences producing significant alignments:                        (Bi
+ts)  Value

gi|89161216|ref|NC_000009.10|NC_000009 Homo sapiens chromosome...  173
+3  0E0

>gi|89161216|ref|NC_000009.10|NC_000009 Homo sapiens chromosome 9, ref
+erence assembly, complete sequence
          Length = 140273252

 Score = 1733 bits (877),  Expect = 0E0
 Identities = 892/897 (99%), Gaps = 1/897 (0%)
 Strand = Plus / Minus

Query: 1        CAAGAAACTATATAGGTATACACTTACGACTTCACAAAACCTATACTTAATATA
+GTAAAT 60
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 94217254 CAAGAAACTATATAGGTATACACTTACGACTTCACAAAACCTATACTTAATATA
+GTAAAT 94217195

Query: 61       CTAAGTAAACATGTATTACTCAAAGTAATATATTTAGAATTATGTATTAGTATA
+AGATCA 120
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 94217194 CTAAGTAAACATGTATTACTCAAAGTAATATATTTAGAATTATGTATTAGTATA
+AGATCA 94217135

Query: 121      GAATTGAATTTAAGTTGTTGGTGACATCTGCATCATTTCATAGGATTAGAACTT
+ACTCAA 180
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 94217134 GAATTGAATTTAAGTTGTTGGTGACATCTGCATCATTTCATAGGATTAGAACTT
+ACTCAA 94217075

Query: 181      AATAATGTAAATCTTTAAAAATATAAATTAGAATGACAAGTGGGAATCATAAAT
+TAAACG 240
                ||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||
Sbjct: 94217074 AATAATGTAAATCTTTAAAAATATAAATTAGAATGACAAGTGGGAATCATAAAT
+TAAACG 94217015

Query= 113 3'_UTR BP;A;C;G;T;OTHER
TATTTTCTTATGTGGGTCTTATGCTTCCATTAACAAATGCTCTGTCTTCAATGATCAAATTTTGAGCAAA
GAAACTTGTGCTTTACCAAGGGGAATTACTGAAAAAGGTGATTACTCCTGAAGTGAGTTTTACACGAACT
GAAATGAGCATGCATTTTCTTGTATGATAGTGACTAGCACTAGACATGTCATGGTCCTCATGGTGCATAT
AAATATATTTAACTTAACCCAGATTTTATTTATATCTTTATTCACCTTTTCTTCAAAATCGATATGGTGG
CTGCAAAACTAGAATTGTTGCATCCCTCAATTGAATGAGGGCCATATCCCTGTGGTATTCCTTTCCTGCT
TTGGGGCTTTAGAATTCTAATTGTCAGTGATTTTGTATATGAAAACAAGTTCCAAATCCACAGCTTTTAC
[download]

I want the output like this:-

Sbjct: 38348698 TTTATTTACAGACCCATAAAAATCCCATCAGTGTCGTCCCCTGCTGAGAGGCCA
+TGTGAG 38348639
Sbjct: 94217074 AATAATGTAAATCTTTAAAAATATAAATTAGAATGACAAGTGGGAATCATAAAT
+TAAACG 94217015
[download]

Thanks in advance
cowboy

Comment on Parsing a file Select or Download Code

Replies are listed 'Best First'.
Re: Parsing a file by jwkrahn (Abbot) on Apr 16, 2009 at 06:52 UTC
`if($_ =~ /Sbjct:\s(\d+)\D+(\d+)$/ .. /\n\n\n\AQuery\=/)` [download] You are using the flip-flop operator so I will assume that there is a single line in `$_`. If so then `/\n\n\n\AQuery\=/` will never match because each line only has one newline and it is at the end of the line. Also there appear to be only two newlines before the string `'Query='` and the anchor `\A` can only match at the beginning of a string so there can't be anything in front of it. You probably want something like this (UNTESTED): `my $subject; while ( <> ) { $subject = $_ if /^Sbjct:\s*\d+\D+\d+$/; print $subject if /^Query=/; }` [download]	[reply] [d/l] [select]
Re: Parsing a file by Anonymous Monk on Apr 16, 2009 at 06:26 UTC
Why not use Bio::Tools::Blast?	[reply]
Re: Parsing a file by citromatik (Curate) on Apr 16, 2009 at 06:40 UTC
I don't know exactly what are you trying to accomplish, but if you only want to extract the "Sbjct" lines you could simply try something like: `while (<$fh>) { print if (/Sbjct/){ }` [download] As a side note, I'm not a fan of bioperl at all, but maybe it would be useful for you. You can try This tutorial to see how to parse Blast reports. Also, it is always better to run blast with the option -m 8 and get the results in tabular format if your intention is to parse the results with a custom made script Hope this helps Update: The code has been updated to meet the poster output sample included citromatik	[reply] [d/l]
Re: Parsing a file by 1Nf3 (Pilgrim) on Apr 16, 2009 at 06:31 UTC
I'm not sure of your desired output. Are you trying to catch the last line containing Sbject? If you would post the desired output for the provided example input file, I'll try to find and post a solution. Anyway, it's very likely that the $_ variable you are parsing contains one line at a time, and searching for multiple \n doesn't work. Regards, Luke	[reply]
Re^2: Parsing a file by cowboyrocks (Novice) on Apr 16, 2009 at 06:51 UTC
Hi.. I have updated my post with the desired output.Thanks	[reply]
Re^3: Parsing a file by 1Nf3 (Pilgrim) on Apr 16, 2009 at 07:15 UTC
If you want to extract Sbjct: lines, and print them out, the following should work fine: `#!/usr/bin/perl -w use strict; while (<>) { print if /^Sbjct:\s*\d+\D+\d+$/; }` [download] After running this code for your sample input data i get: Sbjct: 38348818 AGGCATAAACCACATCCAGCCACCTCCTTCTGATCAGCAGCAAAGCTGACGTTT +TGATCT 38348759 Sbjct: 38348758 CCATCTGTCTGATTCTTGTGTCTACTTCTCAGTTTACAACTCCAGTGGGAAAGA +AAGAGC 38348699 Sbjct: 38348698 TTTATTTACAGACCCATAAAAATCCCATCAGTGTCGTCCCCTGCTGAGAGGCCA +TGTGAG 38348639 Sbjct: 94217254 CAAGAAACTATATAGGTATACACTTACGACTTCACAAAACCTATACTTAATATA +GTAAAT 94217195 Sbjct: 94217194 CTAAGTAAACATGTATTACTCAAAGTAATATATTTAGAATTATGTATTAGTATA +AGATCA 94217135 Sbjct: 94217134 GAATTGAATTTAAGTTGTTGGTGACATCTGCATCATTTCATAGGATTAGAACTT +ACTCAA 94217075 Sbjct: 94217074 AATAATGTAAATCTTTAAAAATATAAATTAGAATGACAAGTGGGAATCATAAAT +TAAACG 94217015 [download] Hope that helps, Luke	[reply] [d/l] [select]
Re: Parsing a file by Anonymous Monk on Apr 16, 2009 at 06:23 UTC
I get `Use of uninitialized value in pattern match (m//) at - line 1.` [download] I think $_ is unefined. Can you show code that actually demonstrates your problem?	[reply] [d/l]