Root Mean Square Inner Product

Calculate the RMSIP between two mode subspaces.

rmsip(...)

# S3 method for enma
rmsip(enma, ncore=NULL, subset=10, ...)

# S3 method for default
rmsip(modes.a, modes.b, subset=10,
                        row.name="a", col.name="b", ...)

Arguments

enma	an object of class "enma" obtained from function nma.pdbs.
ncore	number of CPU cores used to do the calculation. ncore>1 requires package ‘parallel’ installed.
subset	the number of modes to consider.
modes.a	an object of class "pca" or "nma" as obtained from functions pca.xyz or nma.
modes.b	an object of class "pca" or "nma" as obtained from functions pca.xyz or nma.
row.name	prefix name for the rows.
col.name	prefix name for the columns.
...	arguments passed to associated functions.

Details

RMSIP is a measure for the similarity between two set of modes obtained from principal component or normal modes analysis.

Value

Returns an rmsip object with the following components:

overlap

a numeric matrix containing pairwise (squared) dot products between the modes.

rmsip

a numeric RMSIP value.

For function rmsip.enma a numeric matrix containing all pairwise RMSIP values of the modes stored in the enma object.

References

Skjaerven, L. et al. (2014) BMC Bioinformatics 15, 399. Grant, B.J. et al. (2006) Bioinformatics 22, 2695--2696. Amadei, A. et al. (1999) Proteins 36, 19--424.

Author

Lars Skjaerven

See also

pca, nma, overlap.

Other similarity measures: sip, covsoverlap, bhattacharyya.

Examples

if (FALSE) {
# Load data for HIV example
trj <- read.dcd(system.file("examples/hivp.dcd", package="bio3d"))
pdb <- read.pdb(system.file("examples/hivp.pdb", package="bio3d"))

# Do PCA on simulation data
xyz.md <- fit.xyz(pdb$xyz, trj, fixed.inds=1:ncol(trj))
pc.sim <- pca.xyz(xyz.md)

# NMA 
modes <- nma(pdb)

# Calculate the RMSIP between the MD-PCs and the NMA-MODEs
r <- rmsip(modes, pc.sim, subset=10, row.name="NMA", col.name="PCA")

# Plot pairwise overlap values
plot(r, xlab="NMA", ylab="PCA")
}