文本處理-perl

我是生物學家。我需要電腦專家的幫助來幫助編輯文本文件。

這是輸入文件的樣子：

##dsfsd2
##sdf-sdf sasg 5.6.3
gi34_ex Gen  CDS     161     317     .       +       .       Name=Xm ZAK;created by=User
gi56_ex Gen  CDS     2194    2280    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     2848    2951    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     4554    4619    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     4729    4756    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  extracted region        1       11677   .       +       .       Name=Extracted region from gi|371443185|gb|JH556675.1|;Extracted interval="1960862 -> 1972538"

請注意，從第三行開始，輸入有 12 ( $ 1 to $ 12) 列，製表符分隔。文件的最後一行有 16 ( $ 1 to $ 16) 列，製表符分隔。## 行應該被忽略。

我要拿 $ 14 of the last line (interval=“1960862) only the number (1960862) add to $ 4 列 (161,2194,2848..4729) 即 (161+1960862=1961023,2194+1960862=1963056..) 和 $5 列 (317,2280,2951..4756) 即 (317+1960862=1961179, 2280+1960862=1963142)，忽略最後一行。

輸出應如下所示：

##dsfsd2
##sdf-sdf sasg 5.6.3
gi34_ex Gen  CDS     1961023     1961179     .       +       .       Name=Xm ZAK;created by=User
gi56_ex Gen  CDS     1963056    1963142    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     1963710    1963813    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     1965416    1965481    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  CDS     1965591    1965618    .       +       .       Name=Xm ZAK;created by=User
gi37_ex Gen  extracted region        1       11677   .       +       .       Name=Extracted region from gi|371443185|gb|JH556675.1|;Extracted interval="1960862 -> 1972538"

這是一個有效的解決方案。您可以在評論中找到解釋。

#!/usr/bin/perl
use warnings;
use strict;

open my $IN, '<', '1.in' or die $!;

my $line;
$line = $_ while <$IN>;                 # Remember the last line.
my $last = $.;                          # Remember the number of the last line.

my $interval = (split /\t/, $line)[13]; # Extract the 14th column.
$interval =~ s/[^0-9]+//;               # Keep only the number.

seek $IN, 0, 0;                         # Rewind to the beginning of the input.
$. = 0;                                 # Restart the line counter.
my $start = 1;                          # Flag to skip first lines.
while (<$IN>) {
   my @columns = split /\t/;
   /^##/ or undef $start;              # Unset start if the header is over.
   if (not ($start or $. == $last)) {  # Not header or last line?
       $_ += $interval for @columns[3, 4];
   }
   print join "\t", @columns;
}

引用自：https://unix.stackexchange.com/questions/42279