February 11 (Wednesday) -- Stephen Opiyo (Agronomy, Graduate student)

Usage of principal component analysis to model physico-chemical properties of amino acids 
and its applications in bioinformatics


The amount of DNA and amino acid sequences are increasing in databases.  Various methods are 
needed to bring order to these sequences and use them for purposes such as understanding the 
evolutionary relationships between organisms. Multivariate methods have been little used in 
the analysis of protein sequences in bioinformatics.  I will present the use of principal 
component analysis in the modeling of physico-chemical properties of amino acids. I will also 
present its application in the analysis of amino acid sequences. 

Reference:

[1] Lapinsh M., Alexandrs G., Peteris P., Claes P., Torbjorn L., and Jarl E.S. W.  (2002) 
Classification of G-protein coupled receptors by alignment-independent extraction of
pricipal chemical properties of primary amino acid sequences, Protein science 11:795-805.

[2] Wold S., Johnson J., Sjostrom M., Sanberg M., and Rannar S. (1993) DNA and peptide 
sequences and chemical processes multivariately modeled by principal component analysis and 
partial least-squares projections to lantent structures. Anal. Chim. Acta 277: 239-253.