Difference between revisions of "HMM and alignment"

Line 1: Line 1:
 +
* Profile HMMs are similar to simple sequence profiles, but in addition to the amino acid frequencies in the columns of a multiple sequence alignment they contain the position-specific probabilities for inserts and deletions along the alignment
 +
* The logarithms of these probabilities are in fact equivalent to position-specific gap penalties (Durbin et al., 1998).
 +
 +
 
* '''The alignment algorithm maximizes a weighted form of coemission probability, the probability that the two HMMs will emit the same sequence of residues. '''
 
* '''The alignment algorithm maximizes a weighted form of coemission probability, the probability that the two HMMs will emit the same sequence of residues. '''
 
* Amino acids are weighted according to their abundance, rare coemitted amino acids contributing more to the alignment score.  
 
* Amino acids are weighted according to their abundance, rare coemitted amino acids contributing more to the alignment score.  
−
* This weighting is analogous to the use of a null model with amino acid background probabilities in HMM-sequence comparison.
 
  
−
* Profile HMMs are similar to simple sequence profiles, but in addition to the amino acid frequencies in the columns of a multiple sequence alignment they contain the position-specific probabilities for inserts and deletions along the alignment
+
* Secondary structure can be included in the HMM-HMM comparison.
−
* The logarithms of these probabilities are in fact equivalent to position-specific gap penalties (Durbin et al., 1998).  
+
* We score pairs of aligned secondary structure states in a way analogous to the classical amino acids substitution matrices.  
 +
* We use ten different substitution matrices that we derived from a statistical analysis of the structure database, one for each confidence value given by PSIPRED.  
  
  

Revision as of 17:58, 11 February 2013

  • Profile HMMs are similar to simple sequence profiles, but in addition to the amino acid frequencies in the columns of a multiple sequence alignment they contain the position-specific probabilities for inserts and deletions along the alignment
  • The logarithms of these probabilities are in fact equivalent to position-specific gap penalties (Durbin et al., 1998).


  • The alignment algorithm maximizes a weighted form of coemission probability, the probability that the two HMMs will emit the same sequence of residues.
  • Amino acids are weighted according to their abundance, rare coemitted amino acids contributing more to the alignment score.
  • Secondary structure can be included in the HMM-HMM comparison.
  • We score pairs of aligned secondary structure states in a way analogous to the classical amino acids substitution matrices.
  • We use ten different substitution matrices that we derived from a statistical analysis of the structure database, one for each confidence value given by PSIPRED.
Retrieved from "http://wiki.devicerandom.org/index.php?title=HMM_and_alignment&oldid=1371"