Introduction - Sequence Alignment Heuristics

In the second lecture we presented dynamic programming algorithms to calculate the best alignment of two strings. When we are searching, in a database of size 10⁷, for the closest match to a query string of length 200-500, we can not use these algorithms because they require too much time. There are several approaches to overcome this problem:

1.

Implementing the dynamic programming algorithms in hardware, thus executing them much faster.

2.

Using parallel hardware, the problem can be divided efficiently to a number of processors, and the results can be integrated later.

3.

Using different heuristics that work much faster than the original dynamic programming algorithm.

We next present some of the most commonly used heuristics.