Difference: JaysJournal (6 vs. 7)

Revision 72010-05-12 - jayzhang

Line: 1 to 1

META TOPICPARENT	name="NGSAlignerProject"

05/06/10

I'm starting this journal 3 days late (whoops). Here's a brief overview of what happened during my first three days:

Line: 50 to 50

Hopefully get that to pass all the tests (i.e. make the behaviour the same as the non-vectorized version)
Take a look at optimizing for linear gap penalty if I have time.

Added:

>
>

05/12/10

Chris told me that the full alignment was a lower priority, since the score-only alignment is used much more often. So, I've tried optimizing for linear gap penalty instead. I managed to improve performance by not doing a few calculations required in the affine and also improve the memory usage slightly (by not having to use one genome-length-size array).

I did a few preliminary tests, using a randomly generated genome/read sequences of length 1000 each. Over 1000 iterations, my results were:

38.836576365 seconds using affine gap penalty
28.907083597 seconds using linear gap penalty

So I get a roughly 25% decrease in calculation time. Tests using more/fewer iterations and using different genome/read lengths show more or less the same amount.

Now it's on to doing the "banded" algorithm. I'm going to try making a non-vectorized version first, then move on to vectorized. I'm not really sure how I'm going to test it though, since the results are going to be a little bit different...

View topic | History: r73 < r72 < r71 < r70 | More topic actions...