Regelson, Moira Ellen (1997) Protein structure/function classification using hidden Markov models. Dissertation (Ph.D.), California Institute of Technology. http://resolver.caltech.edu/CaltechETD:etd-01142008-095446
Three-dimensional protein structures can be divided into classes in which proteins demonstrate high similarity of structure. While full and accurate determination of a protein's three-dimensional structure from its amino acid sequence is riot feasible at this time, methods seeking to determine this full structure would be aided by a priori information about the sequence's overall structural class. We have utilized Hidden Markov Models on sequences from the SWISS-PROT database in an attempt to determine the structural class of a protein given only its primary amino acid sequence.
Varying representations of the amino acid sequences and the accuracy with which the models using these representations differentiate between classes give some insight into the chemical and physical properties which are significant in the protein folding process. In addition, some representations of the protein sequence can illustrate the redundancy in the protein alphabet and others can capture structural class information with reduced computational requirements. Real vector representations provide an analogy to the problem of speech recognition.
|Item Type:||Thesis (Dissertation (Ph.D.))|
|Degree Grantor:||California Institute of Technology|
|Major Option:||Applied And Computational Mathematics|
|Thesis Availability:||Restricted to Caltech community only|
|Defense Date:||4 June 1997|
|Default Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Imported from ETD-db|
|Deposited On:||01 Feb 2008|
|Last Modified:||04 Mar 2014 19:55|
- Final Version
Restricted to Caltech community only
See Usage Policy.
Repository Staff Only: item control page