home code: covariance  variance  rna graphs  matlash 
[ all code made available under the
GNU Public License ]
covariance component model
matlab code for maximum likelihood estimation of the model described in
Y. Karklin and M. S. Lewicki,
Emergence of complex cell properties by learning to
generalize in natural scenes,
Nature, 2008.
[link]
coming soon... variance component model
matlab code for maximum likelihood estimation of the model described in
Y. Karklin and M. S. Lewicki, A hierarchical Bayesian model for
learning nonlinear statistical regularities in nonstationary natural
signals, Neural Computation, 17 (2): 397423, 2005.
[pdf]
download VarianceComponents.m. (this code implements learning on a synthetic dataset and recovers a set of known components) rna graph classifier
matlab code for classifying noncoding RNAs based on
their secondary structure, as described in
Y. Karklin, R. F. Meraz, S. R. Holbrook.
Classification of
noncoding RNA using graph representations of secondary structure,
Proceedings of the Pacific Symposium on
Biocomputing 10:415, 2005.
[pdf]
download the main package, supplementary files (some more functions for plotting, doing multiclass classification, other utils; highly undocumented!), and a short FAQ. the SVM part of the code was adapted from LSSVMlab, though ultimately SVM training is done with simple matlab matrix division. in order to duplicate results from the paper, you will also need RNAs from RFAM and folding software from ViennaRNA. matlash
a nascent project to convert matlab figures into flash objects.
for example, you can type in matlab,
>> x = linspace(0,2*pi,50);
and get a flash object to embed in a web page:
>> plot(x,cos(x),'r'); hold on; plot(x,sin(2*x),'b'); >> matlash here's how it works:
