Readme file
SERIES C
 Applied Statistics
A hierarchical Bayesian model for predicting the functional consequences of amino-acid polymorphisms, by C.J. Verzilli et al
Journal of the Royal Statistical Society,
Series C, Applied Statistics, Volume
54 (2005) part 1, 191-206
The files lacrepressor.txt and lysozyme.txt contain the data from the mutagenesis experiments used in the paper.
Each column contains native amino acid (AA) and mutant AA-specific information, namely:
aapos = position in chain of native AA
y = effect of mutation on functionality (0==no effect)
ac = solvent accessible area of native AA
rac = accessibility relative to maximum accessibity in data set
rent = normalised phylogenetic entropy of native AA
nrent = phylogenetic entropy of structural neighbourhood of native AA
rb = B-factor of native AA
nrb = B-factor of structural neighbourhood of native AA
uslaa = mutant AA is not in phylogenetic profile
uslby = mutant AA is not in the smallest AA class that includes the phylogenetic profile
bur = mutant AA is charged AA at buried site
trn = mutant AA occurs at glycine or proline in a turn
hlx = mutant AA occurs in helical region and involves glycine or proline
cnsd = native AA is at conserved position in phylogenetic profile
ncnsd = native AA is near conserved position in phylogenetic profile
ifc = native AA is near subunit interface
The gzipped tar archive bayesmars_0.1.0.tar.gz contains the accompanying package "bayesmars" for use in R (http://cran.r-project.org).
The main function `bayesmars' will perform binary classification and regression using Bayesian multivariate adaptive regression splines.
The package has been tested on Linux OS and was prepared using R Version 1.6.2 (10-01-2003). It can be installed using the INSTALL command (see ?INSTALL at the R prompt).
After installation, further information can be obtained typing ?bayesmars or library(help=bayesmars) at the R prompt.
Claudio Verzilli
Department of Epidemiology and Public Health
Imperial College London
St Mary's Campus
Norfolk Place
London
W2 1PG
UK
E-mail: c.verzilli@imperial.ac.uk
http://www1.imperial.ac.uk/med/people/c.verzilli.html
|
