Bioinformatics Algorithms

Multiple Pattern Matching Problem

$n$ = length of text
$m_i$ = length of each pattern
$p$ = number of patterns

Bruteforce Algorithm

$$ O(n_m_1+ n_m_2 + ... + n_m_p) \\ = O(n_\sum_{i=1}^{p}m_i) $$

KMP Algorithm

$$ O( (n+m_1) + (n+m_2) + ... + (n+m_p) ) \\ = O(n*p + \sum_{i=1}^{p}m_i) $$

Trie

Preprocessing for calculating Trie(patterns)

$$ O(\sum_{i=1}^{p}m_i) $$

Pattern Matching

$$ O(n * max(m_i)) $$

Finally, the total time complexity is

$$ O(\sum_{i=1}^{p}m_i + n * max(m_i)) $$

Suffix Trie

Preprocessing for calculating SuffixTrie(text) using all the suffixes of the text

$$ O(n^2) $$

Pattern Matching

$$ O(\sum_{i=1}^{p}m_i) $$

Finally, the total time complexity is

$$ O(n^2 + \sum_{i=1}^{p}m_i) $$

Suffix Array

Preprocessing for calculating SuffixArray(text)

$$ O(n * log(n)) $$

Pattern Matching

$$ O(p * log(n)) $$

Finally, the total time complexity is

$$ O(n * log(n) + p * log(n)) \\ = O( (n+p) * log(n)) $$

greatly reduced memory complexity

Motif Finding

TBD