Skip to content

Latest commit

 

History

History
131 lines (113 loc) · 4.32 KB

R-tips-and-tricks.md

File metadata and controls

131 lines (113 loc) · 4.32 KB
Raw
title description category subcategory tags
R tips
This code helps with regular data improving efficiency
computing
tips_tricks
R
visualization

Import/Export of files

Stop using write.csv, write.table and use the rio library instead. All rio needs is the file extension to figure out what file type you're dealing with. Easy import and export to Excel files for clients.

Parsing in R using Tidyverse

This is a link to a nice tutorial from Ista Zahn from IQSS using stringr and tidyverse for parsing files in R. It is from the Computefest 2017 workshop: http://tutorials-live.iq.harvard.edu:8000/user/zwD2ioESyGbS/notebooks/workshops/R/RProgramming/Rprogramming.ipynb

Better clean default ggplot

install cowplot (https://cran.r-project.org/web/packages/cowplot/index.html)

library(cowplot)

Nice looking log scales

Example for x-axis

library(scales)
   p + scale_x_log10(
         breaks = scales::trans_breaks("log10", function(x) 10^x),
         labels = scales::trans_format("log10", scales::math_format(10^.x))) +
       annotation_logticks(sides='b')

Read a bunch of files into one dataframe

library(tidyverse)
read_files = function(files) {
  data_frame(filename = files) %>%
    mutate(contents = map(filename, ~ read_tsv(.))) %>%
    unnest()
}

remove a layer from a ggplot2 object with ggedit

plotGeneSaturation(bcb, interestingGroups=NULL) +
  ggrepel::geom_text_repel(aes(label=description, color=NULL))
p %>%
  ggedit::remove_geom('point', 1) +
  geom_point(aes(color=NULL))

Link to information about count normalization methods

The images currently break, but I will update when the course materials are in a more permanent state.

.Rprofile usefulness

## don't ask for CRAN repository
options("repos" = c(CRAN = "http://cran.rstudio.com/"))
## for the love of god don't open up tcl/tk ever
options(menu.graphics=FALSE)
## set seed for reproducibility
set.seed(123456)
## don't print out more than 100 lines at once
options(max.print=100)
## helps with debugging Bioconductor/S4 code
options(showErrorCalls = TRUE, showWarnCalls = TRUE)

## Create a new invisible environment for all the functions to go in
## so it doesn't clutter your workspace.
.env <- new.env()

## ht==headtail, i.e., show the first and last 10 items of an object
.env$ht <- function(d, n=10) rbind(head(d, n), tail(d, n))

## copy from clipboard
.env$pbcopy = function(x) {
  capture.output(x, file=pipe("pbcopy"))
}

## update your local bcbioRNASeq and bcbioSingleCell installations
.env$update_bcbio = function(x) {
    devtools::install_github("steinbaugh/basejump")
    devtools::install_github("hbc/bcbioBase")
    devtools::install_github("hbc/bcbioRNASeq")
    devtools::install_github("hbc/bcbioSingleCell")
}

attach(.env)

Make density plot without underline

ggplot(colData(sce) %>%
       as.data.frame(), aes(log10GenesPerUMI)) +
    stat_density(geom="line") +
    facet_wrap(~period + intervention)

Archive a file to Dropbox with a link to it

```{r results='asis'}
dropbox_dir = "HSPH/eggan/hbc02067"
archive_data_with_link = function(data, filename, description, dropbox_dir) {
    readr::write_csv(data, filename)
    links = bcbioBase::copyToDropbox(filename, dropbox_dir)
    link = gsub("dl=0", "dl=1", links[[1]]$url)
    basejump::markdownLink(filename, link, paste0(" ", description))
}
archive_data_with_link(als, "dexseq-all.csv", "All DEXSeq results", dropbox_dir)
archive_data_with_link(als %>%
                     filter(padj < 0.1), "dexseq-sig.csv",
                     "All significant DEXSeq results", dropbox_dir)

Novel operators from magrittr

The “%<>%” operator lets you pipe an object to a function and then back into the same object. So: foo -> foo %>% bar() is the same as foo %<>% bar()

gghelp: Converts a natural language query into a 'ggplot2' command

This package allows users to issue natural language commands related to theme-related styling of plots (colors, font size and such), which then are translated into valid 'ggplot2' commands.

Examples:

gghelp("rotate x-axis labels by 90 degrees")
gghelp("increase font size on x-axis label")
gghelp("set x-axis label to 'Length of Sepal'")