I spent the first half of this week debugging the perl code I wrote last week. Unfortunately, many of the protein sequences I was given to compare were slightly inconsistent, and the variations seemed to follow several different patterns. I talked it over with Brian, and we ended up deciding which incosistencies the program should ignore and which the programs should consider erroneous. It took the first half of the week to get all the details worked out and then coded.
Midway through the week Brian and I met with Dr. Kavraki to go over our goals for the summer. By the end of the summer she would like us to
1.) Improve the performance and correctness of the current algorithm
2.) Improve the running time and implement a hierarchtical pattern for the algorithm
3.) Improve the quality of results while allowing for more fuzziness and recognize motifs with increased tolerance
4.) Brian is supposed to write a bibliography including previous work
The second half of the week was spent writing more code. Brian asked me to write a make file that would run his geoHash program and then my perl program on different benchmark proteins to make sure that the programs are still working correctly. With these tests we can make changes to the code without having to go through a tedious error checking process to make sure we did cause any errors. I created three tests: one for proteins that should work correctly, one for proteins that should fail, and another where the user can define which proteins he would like to test. I spent the rest of the week learning to write a make file then writing more perl code to get the tests to work properly.