cparsania / tidywrappers

An R package to deal with an object tbl

TidyWrappers

TidyWrappers is an R package to deal with an object tbl. It provides handy wrapper functions on top of dplyr verbs to get, count, convert, keep, remove and replace data from a tbl object.

Motivation

The way R is being used, has changed drastically in last few years due to the availability of tidyverse and more specifically dplyr (all credits to Hadley Wickham and RStudio team). Rich documentation, stable release and excellent functionalities of these packages have provoked significant number of R users to switch from traditional dataframes to intelligible data structure, tibble (aka tbl). dplyr provides users with rich utilities like select, mutate, filter, group_by , summarise and arrange for easy manipulation of a tbl. However, for newbies it takes a while to become familiar with tidy data and dplyr core and helper functions.

TidyWrappers contains set of functions, which save you little from writing and implementing core dplyr verbs while manipulating data from tbl. Additionaly, redundant lines of code can be avoided using functions given in TidyWrappers. Current version of TidyWrappers containes more than 30 functions, which are wrapped on top of various dplyr functions. See below in examples.

Install

Install current version of TidyWrappers on your system.

# install.packages("devtools")
library(devtools)
devtools::install_github("cparsania/TidyWrappers", dependencies = TRUE)

Examples

library(dplyr)
library(TidyWrappers)
library(magrittr)

  tbl <- tibble::tibble(a = letters[1:6], b = NA_character_, x = c(0,1,2,0,0,NA) , y = c(0,1,2,0,3,5) , z = c(0,1,2,0,3,5) )
  tbl
#> # A tibble: 6 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 a     <NA>      0     0     0
#> 2 b     <NA>      1     1     1
#> 3 c     <NA>      2     2     2
#> 4 d     <NA>      0     0     0
#> 5 e     <NA>      0     3     3
#> 6 f     <NA>     NA     5     5
  
  ## keep rows having  0 across all numeric columns. 
  
  # using dplyr
  tbl %>% dplyr::filter_if(is.numeric , dplyr::all_vars( . == 0) )
#> # A tibble: 2 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 a     <NA>      0     0     0
#> 2 d     <NA>      0     0     0
  
  # using TidyWrappers
  tbl %>% TidyWrappers::tbl_keep_rows_zero_all()
#> # A tibble: 2 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 a     <NA>      0     0     0
#> 2 d     <NA>      0     0     0

  ## remove rows having  0 across all numeric columns. 
  
  # using dplyr 
  tbl %>% dplyr::filter_if(is.numeric , dplyr::any_vars( . > 0) )
#> # A tibble: 4 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 b     <NA>      1     1     1
#> 2 c     <NA>      2     2     2
#> 3 e     <NA>      0     3     3
#> 4 f     <NA>     NA     5     5
  
  # using TidyWrappers 
  tbl %>% TidyWrappers::tbl_remove_rows_zero_all()
#> # A tibble: 4 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 b     <NA>      1     1     1
#> 2 c     <NA>      2     2     2
#> 3 e     <NA>      0     3     3
#> 4 f     <NA>     NA     5     5
  
  ## Convert log2 to all numeric columns optionally adding numeric fraction to original values. 
  
  # using dplyr 
  tbl %>% dplyr::mutate_if(is.numeric , ~ log2 (. +  1))
#> # A tibble: 6 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 a     <NA>   0     0     0   
#> 2 b     <NA>   1     1     1   
#> 3 c     <NA>   1.58  1.58  1.58
#> 4 d     <NA>   0     0     0   
#> 5 e     <NA>   0     2     2   
#> 6 f     <NA>  NA     2.58  2.58
 
  # using TidyWrappers
  tbl %>% TidyWrappers::tbl_convert_log2(frac = 1)
#> # A tibble: 6 x 5
#>   a     b         x     y     z
#>   <chr> <chr> <dbl> <dbl> <dbl>
#> 1 a     <NA>   0     0     0   
#> 2 b     <NA>   1     1     1   
#> 3 c     <NA>   1.58  1.58  1.58
#> 4 d     <NA>   0     0     0   
#> 5 e     <NA>   0     2     2   
#> 6 f     <NA>  NA     2.58  2.58

  
  ## Remove columns / variables  having atleaset one NA
  
  # using dplyr
  tbl %>% dplyr::select_if( ~ ( (is.na(.)) %>% any(., na.rm = T)  %>% `!`) ) 
#> # A tibble: 6 x 3
#>   a         y     z
#>   <chr> <dbl> <dbl>
#> 1 a         0     0
#> 2 b         1     1
#> 3 c         2     2
#> 4 d         0     0
#> 5 e         3     3
#> 6 f         5     5
  
  # using TidyWrappers
  tbl %>% TidyWrappers::tbl_remove_vars_NA_any()
#> # A tibble: 6 x 3
#>   a         y     z
#>   <chr> <dbl> <dbl>
#> 1 a         0     0
#> 2 b         1     1
#> 3 c         2     2
#> 4 d         0     0
#> 5 e         3     3
#> 6 f         5     5
  
  ## Remove columns / variables  having all values are NA
  
  # using dplyr
  tbl %>% dplyr::select_if( ~ ( (is.na(.)) %>% all(., na.rm = T)  %>% `!`) ) 
#> # A tibble: 6 x 4
#>   a         x     y     z
#>   <chr> <dbl> <dbl> <dbl>
#> 1 a         0     0     0
#> 2 b         1     1     1
#> 3 c         2     2     2
#> 4 d         0     0     0
#> 5 e         0     3     3
#> 6 f        NA     5     5
  
  # using TidyWrappers
  tbl %>% TidyWrappers::tbl_remove_vars_NA_all()
#> # A tibble: 6 x 4
#>   a         x     y     z
#>   <chr> <dbl> <dbl> <dbl>
#> 1 a         0     0     0
#> 2 b         1     1     1
#> 3 c         2     2     2
#> 4 d         0     0     0
#> 5 e         0     3     3
#> 6 f        NA     5     5

Reference

Follow page Reference to see full list of functions.