calc.diffusion.kernel {GOSim} | R Documentation |
Manifold embeddings of gene ontology terms via diffusion kernel techniques. Diffusion kernels are positive semidefinite similarity measures calculated from the graph Laplacian. They are interpreted as the result of a local heat diffusion process along the graph structure.
calc.diffusion.kernel(method="diffKernelLapl", m=7) load.diffusion.kernel(method="diffKernelLapl")
method |
one of "diffKernelLapl", "diffKernelpower", "diffKernelLLE", "diffKernelexpm" |
m |
(1) Half the power of the transition probability matrix (an integer > 0). (2) an arbitrary positive time constant for the exponential diffusion kernel |
calc.diffusion.kernel
puts a kernel matrix / similarity matrix named "<method><ontology><organism><evidence levels>.rda" in the current directoy. This file has then to be moved into the data directory of
GOSim. It can be used afterwards by calling load.diffusion.kernel
.
Sometimes it is necessary to switch to the GOSim directory for this purpose.
Lerman G. & Shaknovich B., Defining Functional Distance using Manifold Embeddings of Gene Ontology Annotations, PNAS, 104(27): 11334 - 11339, 2007
## Not run: calc.diffusion.kernel(method="diffKernelpower", m=5) load.diffusion.kernel("diffKernelpower") getTermSim(c("GO:0007166","GO:0007267","GO:0007584","GO:0007165","GO:0007186"),method="diffKernel",verbose=FALSE) ## End(Not run) # --> this may take some time ...