How to delete all the lines that match specific condition - bash

Question

I have a number of pdb files and I want to keep only those lines that starts with ^FORMUL and if line has C followed by the number that is larger then (C3,C4,C5,C6..100 etc) then I should not print it. Second condition is that within every line sum of C, H and N should be >= 6, then delete it
So overall delete the lines in which C is followed by number more then 2> and sum of C+O+N is >= then 6.
FORMUL 3 HOH *207(H2 O) (print it)
FORMUL 2 SF4 FE4 S4 (print it)
FORMUL 3 NIC C5 H7 N O7 (don't print, there is C5 and sum is more then 6)
FORMUL 4 HOH *321(H2 O) (print it)
FORMUL 3 HEM 2(C34 H32 FE N4 O4) (don't print, there is C34)
I have tried to do it in perl but lines are soo diverse from each other so Im not sure if it is possible to do.
Over all these conditions chould be included together, meaning that all lines in which C>2 and sum>=6 should be deleted.
C1 O5 N3 should be deleted; C3 N1 01 should not be deleted although C is 3.
In perl I don't know how to assign these two conditions. Here I wrote opposite situation not to delete but to print these lines if these two conditions are not met.
#!/usr/bin/perl
use strict;
use warnings;
my #lines;
my $file;
my $line;
open ($file, '<', '5PCZ.pdb') or die $!;
while (my $line = <$file>)
{
if ($line =~ m/^FORMUL/)
{
push (#lines, $line);
}
}
close $file;
#print "#lines\n";
foreach $line(#lines)
{
if ($line eq /"C"(?=([0-2]))/ )
{
elsif ($line eq "Sum of O,N & C is lt 6")
print #lines
}
}

Answer 1

As you've seen, it's probably easier to write this as a filter that prints the lines that you want to keep. I've also written this following the Unix Filter Model (reads from STDIN, writes to STDOUT) as that makes the program far more flexible (and, interestingly, easier to write!)
Assuming that you're running the program on Linux (or similar) and that your code is in an executable file called my_filter (I recommend a more descriptive name!) then you would call it like this:
$ my_filter < 5PCZ.pdb > 5PCZ.pdb.new
The code would look like this:
#!/usr/bin/perl
use strict;
use warnings;
while (<>) { # read from STDIN a line at a time
# Split data on whitespace, but only into four columns
my #cols = split /\s+/, $_, 4;
next unless $cols[0] eq 'FORMUL';
# Now extract the letter stuff into a hash for easy access.
# We extract letters from the final column in the record.
my %letters = $cols[-1] =~ m/([A-Z])(\d+)/g;
# Give the values we're interested in, a default of 0
$letters{$_} //= 0 for (qw[C O N]);
next if $letters{C} > 2
and $letters{C} + $letters{O} + $letters{N} >= 6;
# I think we can then print the line;
print;
}
This seems to give the correct output for your sample data. And I hope the comments make it obvious how to tweak the conditions.

Answer 2

Extended Awk solution:
awk -F'[[:space:]][[:space:]]+' \
'/^FORMUL/{
if ($4 !~ /\<C/) print;
else {
match($4, /\<C[0-9]+/);
c=substr($4, RSTART+1, RLENGTH);
if (c > 2) next;
else {
match($4, /\<O[0-9]+/);
o=substr($4, RSTART+1, RLENGTH);
match($4, /\<N[0-9]+/);
n=substr($4, RSTART+1, RLENGTH);
if (c+o+n < 6) print
}
}
}' 5PCZ.pdb
The output:
FORMUL 3 HOH *207(H2 O)
FORMUL 2 SF4 FE4 S4
FORMUL 4 HOH *321(H2 O)