META-MEME

The suite of tools described in this section enables the discovery of motifs in groups of related protein or DNA sequences (MEME), the use of these motifs to search a sequence database (MAST), and the integration of these motifs into an HMM 

MEME [91] uses statistical modeling (expectation maximization) techniques to construct conserved sequence motifs from a collection of, presumably, related protein or nucleic acid sequences. According to the Web site, 

MEME represents motifs as position-dependent letter-probability matrices which describe the probability of each possible letter at each position in the pattern. Individual MEME motifs do not contain gaps. Patterns with variable-length gaps are split by MEME into two or more separate motifs. MEME takes as input a group of DNA or protein sequences (the training set) and outputs as many motifs as requested. MEME uses statistical modeling techniques to automatically choose the best width, number of occurrences, and description for each motif, (http //meme.sdsc.edu/meme/intro.html) 

MAST [92] uses a motif description (e.g., one generated by MEME) to search a sequence database for sequences matching the motif. 

META-MEME [93-95] can be used to integrate motifs discovered by MEME, as well as information derived from sets of related sequences, into a single motif-based HMM. The latter can be used to search sequence databases for homologs. 

---

MEME, MAST, and META-MEME are important enough to be covered separately in a later section. 

Grundy, W. N., T. L. Bailey, C. P. Elkan, and M. E. Baker. 1997. Meta-MEME Motif-based hidden Markov models of biological sequences. Comput App Biosci 13 397-406. 

---