DNBr

This code is written by R, to compute the Dynamic Network Biomarkers(DNB) model developed by ChenLab in CAS, China.

If you're interested in this subject, please

• go to http://sysbio.sibcb.ac.cn/sysbio/cb/chenlab/LuonanChen.htm or
• email lnchen@sibs.ac.cn to let us know.

To install this package, use the following command in R:

devtools::install_github("Kaiyu-W/DNBr")

More details see the codes by vignette("DNBr") or ?DNBr::function in R, please.

Log:

• Add choice high_cutoff = -1 for DNBcompute() to skip selecting highly variable genes.

• Add parameter force_allgene for DNBfilter() to control which genes are used as background.

• Note that,

  • To stay consistent with DNBcompute(), DNBfilter(force_allgene = FALSE) as default.
  • There may be some missing genes when DNBfilter(force_allgene = FALSE) because of different high variable genes among groups, and those missing ones will later be removed from modules during computation. Use DNBcompute(..., high_cutoff = -1) to avoid this.
  • Use DNBfilter(..., force_allgene = TRUE) to re-produce the output by early version.

Example:

library(DNBr)
data(data.example)
data(meta.example)

a <- DNBcompute(data.example, meta.example)
b <- DNBfilter(a, ntop = 5, force_allgene = TRUE)
DNBplot(b, show = TRUE, save_pdf = FALSE)