BioPhysConnectoR (version 1.6-5)

simc: Computed Elastic Network Models for Switched-Off-List of Contacts

Description

For each entry in the contact list the contact will be broken and the resulting covariance matrix and new B factors will be computed in the elastic network model. Furthermore the Frobenius norms between the original and the new covariance matrix can be evaluated.

Usage

simc(pdb, mj1 = NULL, mj2 = NULL, mj.avg = FALSE, cl = NULL,
     alpha = 82, cuts = 169, path = getwd(), inv2file = FALSE,
     bfacs = TRUE, frob = TRUE, loc = NULL, norm = FALSE,
     file = NULL, cluster = NULL)

Arguments

pdb

file name of the PDB

mj1

matrix for the intrachain interaction strengths

mj2

matrix for the interchain interaction strengths

mj.avg

logical, if TRUE only the average value of the matrices is used as value for the interaction of any two amino acids

cl

optional contact list to process

alpha

strength of the peptide bond

cuts

squared cutoff distance

path

path to the output files

inv2file

logical, if TRUE the inverse Hessian matrix is written to a file, otherwise it will not be stored

bfacs

logical, if TRUE, the B factors are written to a file, otherwise they will not be stored

frob

logical, if TRUE, the Frobenius norm is computed

loc

dimensions i1, j1, i2, j2 for a matrix subset of which the frobenius norm should be computed

norm

logical, if TRUE the Frobenius norm is computed for the normalized matrices

file

personalized file name prefix

cluster

snow cluster object created with makeCluster()

Value

• No values are returned.

Details

If no contact list is given, the full contact list is extracted from the PDB-file. Each contact (except covalent contacts) in the list is broken and the corresponding covariance matrix and B factors are computed. Those can be written into files. A user-defined contact list can be specified as well. For the computation of the Frobenius norm, different regions can be specified in loc as matrix. Each row determines a region [i1:j1,i2:j2] to be used for the norm. The routine is parallelized for the list of contacts using parLapply() from the package snow. If cluster is left at its default value the computaion is carries out in serial.

References

Hamacher and McCammon (2005) Journal of Chemical Theory and Computation 2, 873. Hamacher (2008) Gene 422, 30--36. Tierney, Rossini, Li (2009) Int J Parallel Proc 37, 78--90.

See Also

sim, sims

Examples

cl<-matrix(c(3,1,4,1,5,1,9,1,10,1,11,1,24,1,66,1,67,1,68,1),ncol=2,byrow=TRUE)
out<-simc(system.file("1KZK.pdb", package = "BioPhysConnectoR"), cuts = 169, cl=cl)

## Cluster example
makeCluster(2)->clu
out<-simc(system.file("1KZK.pdb", package = "BioPhysConnectoR"), cuts = 169, cl=cl, 
                  cluster=clu)
stopCluster(clu)