Sstrhsrtj

I have a large file that contains 4 columns and 7,000 lines. I need to remove the rows in which the start of the second column is the same as the start to the fourth column.

Input:

Gator_locus75   AATTCCATGTACG   Gator_locus23   CTAGAGGAAGT

Gator_locus18   AATTCCATTATGG   Gator_locus14   AATTCAAAAAAT

Gator_locus13   CTAGAACCCACC    Gator_locus72   CTAGAATGTATG

Gator_locus16   AATTCATCCTCT    Gator_locus15   CTAGATTGCCAA

Gator_locus24   CTAGAGCTGCTG    Gator_locus12   AATTCAGTCCAC

Output:

Gator_locus75   AATTCCATGTACG   Gator_locus23   CTAGAGGAAGT

Gator_locus16   AATTCATCCTCT    Gator_locus15   CTAGATTGCCAA

Gator_locus24   CTAGAGCTGCTG    Gator_locus12   AATTCAGTCCAC

I need to remove the rows in which the string the the second column starts "AATT" and the string in the same row fourth column starts "AATT". I also need to do the same thing when the string in the second column starts "CTAG" and the string in the fourth column starts "CTAG".

To print lines where the first 4 characters of column 2 are not equal to the first 4 characters of column 4:

awk 'substr($2, 1, 4) != substr($4, 1, 4)' < input

This uses the main code as a "test" to see whether a line should be printed; there's no explicit action section, since the default-print action is what we want. The main code simply extracts the first four characters from each column and compares them.

To remove rows in which the second field starts with AATT and the fourth field starts with AATT, and the same with CATG:

awk '($2 !~ /^AATT/ && $4 !~ /^AATT/) && ($2 !~ /^CTAG/ && $4 !~ /^CTAG/) {print}' /path/to/file

As a more general solution:

awk 'substr($2,1,4) != substr($4,1,4) {print}' /path/to/file