BI_BalanceSplit: split a data frame into balanced parts according to specified...

In shijianasdf/BasicBioinformaticsAnalysisFromZhongShan: Launch Usual Clinical tool for Kind Yield

Description

BalanceSplit help quickly splitting a data frame into balanced parts according to specified cluster factor.

Usage

  BalanceSplit(data,
             cluster,
             nsplit = 2,
             seed.range=1:500,
             p.cutoff=0.1)
   summary.BalanceSplit(BalanceList,
                        upset.seed=c(3,11),
                        set=1)

Arguments

data	a data frame containing cluster cols.
cluster	cluster colnames.
nsplit	the number of balance splitting.Default is 2.
seed.range	candidate seeds.Default is 1:500.
p.cutoff	the cut off of P value.Default is 0.1.

Author(s)

Weibin Huang<[e-mail]>

Examples

library(lucky)
Plus.library(c("tableone","knitr"))
data(data.TableOne)
data=data.TableOne
cluster = c("age","gender","his1","his2","his.grade","Stage","pT","N.status","M.status","chem.status")
L <- BalanceSplit(data,
                  cluster,
                  nsplit = 2,
                  seed.range=1:500,
                  p.cutoff=0.1)
table1 <- L[["seed_456"]][["table"]]
kable(table1[,1:3],align = 'c')
data.TableOne2 <- L[["seed_456"]][["data"]]

## use upset plot to visualize the difference between seeds
summary.BalanceSplit(BalanceList = L,
                     upset.seed=c(3,11),
                     set=1)

shijianasdf/BasicBioinformaticsAnalysisFromZhongShan documentation built on Jan. 3, 2020, 10:08 p.m.