Intermediate programming with R

Reference

Explanation of columns in “counts-raw.txt.gz”:

• doi - digital object identifier
• pubDate - Date of publication, format: yyyy-mm-dd
• journal - Abbreviation of PLOS journal
• title - Title of article
• articleType - Article classification
• authorsCount - Number of authors
• f1000Factor - Score assigned by Faculty of 1000
• backtweetsCount - Number of tweets
• deliciousCount - Number of bookmarks in Delicious
• pmid - PubMed ID number
• plosSubjectTags - Descriptions of the subject areas of the article
• plosSubSubjectTags - More specific descriptions of the subject areas of the article
• facebookShareCount - Number of Facebook shares
• facebookLikeCount - Number of Facebook likes
• facebookCommentCount - Number of Facebook comments
• facebookClickCount - Number of Facebook clicks
• mendeleyReadersCount - Number of Mendeley users that bookmarked article
• almBlogsCount - Number of blog posts that link to article
• pdfDownloadsCount - Number of PDF downloads
• xmlDownloadsCount - Number of XML downloads
• htmlDownloadsCount - Number of page views
• almCiteULikeCount - Number of saves in CiteULike
• almScopusCount - Number of citations in Scopus
• almPubMedCentralCount - Number of citations in PubMed Central
• almCrossRefCount - Number of citations in CrossRef
• plosCommentCount - Number of comments on PLOS website
• plosCommentResponsesCount - Number of responses to comments on PLOS website
• wikipediaCites - Number of links to article
• year - Year of publication, format: yyyy
• daysSincePublished - Number of days since publication (out-dated)
• wosCountThru2010 - Number of citations in Web of Science as of 2010
• wosCountThru2011 - Number of citations in Web of Science as of 2011

dplyr

• filter - subset rows
• select - subset columns
• arrange - sort column(s)
• mutate - create new column(s)
• group_by - split data into groups based on values in column(s)
• summarize - reduce all rows (per group) to one summary row
• %>% - pipe output of one function to the next

ggplot2

• aes - Map columns of data frame to plot aesthetics
  • x - data on x-axis
  • y - data on y-axis
  • color - color of points and lines
  • shape - shape of points
  • size - size of points
  • fill - color of geometric shapes
• geom_* - The geometric objects to be plotted
  • geom_point - scatter plot
  • geom_bar - bar plot
  • geom_histogram - histogram
  • geom_smooth - loess curve
  • geom_text - use text labels instead of points
  • geom_errorbar - Add error bars
• scale_x_log10, scale_y_log10 - Log transform an axis
• scale_x_continuous, scale_x_discrete - Change breaks and labels on axis
• scale_color_manual, scale_fill_manual, scale_color_brewer - Change colors used for color or fill aesthetics
• facet_grid, facet_wrap - Create a plot per group
• theme - Change the appearance of the plot

Debugging

• debug(function_name) - Enter debugger whenever function is called
• browser() - Enter debugger
• options(error = recover) - Set this option to enter debugger whenever an error occurs

Defensive programming

• stopifnot(cond1, cond2, …) - Stop if any of the listed conditions are FALSE
• From package testit:
  • assert(“message”, cond1, …) - Stop and print message if any of the listed conditions are FALSE
  • has_warning(expr) - Return TRUE if expression creates a warning
  • has_error(expr) - Return TRUE if expression creates an error