Not quite sure what you are asking since your data isn't formatted, put <code> and </code> tags around the data and it will appear as you expect.

I'll guess that you want to collect whole lines that have a certain word

my @lines;
while ( my $line = <$input_fh> ) {
    push @lines, $line if $line =~ /WORD/;
}
[download]

use strict;
use warnings;

my @all_lines = <DATA>;

# get the lines that match the word 'number'
my @lines = grep /number/, @all_lines;

print "Lines that matched\n";
print for @lines;

# get and highlight the lines that match the word 'number'
my @bold = map { $_ =~ /number/ and s/number/<b>number<\/b>/g ? $_ : (
+); } @all_lines;

print "\nHighlighted word\n";
print for @bold;

__DATA__
number....................
.....................
......number..............
.........................
..................number
[download]

outputs:

Lines that matched
number....................
......number..............
..................number

Highlighted word
<b>number</b>....................
......<b>number</b>..............
..................<b>number</b>
[download]

also is it possible, in the result the word searched to be displayed in bold

That depends on how the result is displayed. If the output goes into a terminal, you can use Term::ANSIColor for the formatting, other choices of output require other things to do.

As you can see your questions are rather vague, so the answers must be as well.

When you search for a regex, you can match it against several elements in the string. Assuming you mean a literal number, rather than the word "number", use m/^(\d+)(.+)$/, as it will only match if there's a number at the very start.

Once matched, the number itself will be assigned to $1, and the rest of the line, up to the EOL mark, to $2.
Make sure to match only what you need, though. If the number is displayed as 'xxxx.xx', or 'xx,xxx', the regex will fail.
As for the second part, yes, it is possible. How is a question of the output format, so more information is needed.