influence_calc: InDePTH main function to calculate hub score.

In koido/InDePTH: influential gene detection in perturbed transcriptome hierarchical network

Description

This function constructs DEGs network showing the degree of influence and from the network, calculate hub score. This function use the result object of cscore_calc function.

Usage

 influence_calc(up.sig, dn.sig, up.ratio, dn.ratio, tot.thr, input.dir, output.dir, cscore = res, write.name = NULL)

Arguments

up.sig	character vector of up sigunature of gene sets (Probe ID of Human Genome U133 Plus 2.0 Array is acceptable).
dn.sig	character vector of down sigunature of gene sets (Probe ID of Human Genome U133 Plus 2.0 Array is acceptable).
up.ratio	numeric vector showing changes of gene expression levels of up sigunature of gene sets (probe id or official gene symbol).
dn.ratio	numeric vector showing changes of gene expression levels of down sigunature of gene sets (probe id or official gene symbol).
tot.thr	numeric of threshold to select genetic perturbation from the result of cscore_calc function.
input.dir	a character string naming a directory containing the result of init function.
output.dir	a character string naming a directory for output files from this function.
cscore	a object from cscore_calc function.
write.name	a character string naming a file to write network plot in igraph object (RDS format) and GML format. Suffix is NOT required.

Author(s)

Masaru Koido ([e-mail])

References

M. Koido, et al., 'Detection of hub genes for developing gene expression networks under anticancer drug treatment', Oncotarget, 9(49), 29097-29111, 2018.

http://www.lincsproject.org/

Examples

## Not run:
#-- only the first time of InDePTH
# init(input.dir = "input", out.dir = "init")
#
# Run (arbitrary threshold)
res1 <- cscore_LINCS(up.sig = c( "217761_at", "217398_x_at", "218744_s_at" ),
                     dn.sig = c( "221856_s_at", "218193_s_at", "218457_s_at" ),
                     input.dir = "init",
                     output.dir = "out",
                     write.name = NULL)
res2 <- influence_calc(up.sig = c( "217761_at", "217398_x_at", "218744_s_at" ),
                      dn.sig = c( "221856_s_at", "218193_s_at", "218457_s_at" ),
                      up.ratio = c( 2, 4, 6 ),
                      dn.ratio = c( 1/2, 1/3, 1/5 ),
                      tot.thr = 0.5,
                      input.dir = "init",
                      output.dir = "out",
                      cscore = res1,
                      write.name = NULL)

# Run (Data-driven threshold)
thr_res <- make_threshold( pert_id_vec = c( "TRCN0000010389", "TRCN0000010390", "TRCN0000010391"),
                           cscore = res1,
                           cell = "HT29",
                           input.dir = "init",
                           output.dir = "out" )
res2 <- influence_calc(up.sig = c( "217761_at", "217398_x_at", "218744_s_at" ),
                      dn.sig = c( "221856_s_at", "218193_s_at", "218457_s_at" ),
                      up.ratio = c( 2, 4, 6 ),
                      dn.ratio = c( 1/2, 1/3, 1/5 ),
                      tot.thr = thr_res$cutoff,
                      input.dir = "init",
                      output.dir = "out",
                      cscore = res1,
                      write.name = NULL)

## End(Not run)

koido/InDePTH documentation built on May 28, 2019, 9:50 a.m.