Connecting R to a SQL Server can be achieved using the RODBC or RJDBC libraries. Under CentOS this is a little more challenging than Windows and is generally achieved by installing the freeTDS ODBC driver and using the RODBC library. I found compilation a little difficult under CentOS 5.7 and so I moved on to JDBC …

Read More

Introduction I’m a big fan of the R packages in the tidyverse and dplyr in particular for performing routine data analysis. I'm currently working on a project that requires fitting dose-repsonse curves to data. All works well when the amount of data is small but as the dataset grows so does the computational time. …

Read More

R is a very powerful language but was designed to run single-threaded. There are libraries that will run parallel code but there is no inherent support to run a background process and monitor for completion in the same way that you can in, for example, C++. One way to overcome this is to launch a second instance of R …

Read More

queryBuilder is an htmlwidget wrapper for jQuery queryBuilder, a fantastic jQuery plugin to create queries and filters. The htmlwidget generates a dplyr string that can be used to filter a data frame within a shiny app. In addition I'm playing with some experimental features to expand functionality. The first is a way …

Read More

I’m starting to experiment with RStudio Connect. I’ve been working on a shiny app that writes to a database and a flexdashboard that can illustrate the output. When writing this type of approach using Shiny Server, both pieces of code could simply point to the same sqlite database. This becomes a little more …

Read More

There seems to be a fad right now of merchandising articles containing words made up of chemical symbols. Beryllium Erbium (BeEr) is particularly popular and I have to admit I have a T-Shirt claiming "Call me NErDy". I've put together a small code in R which will check to see if a word can be created purely …

Read More

There are times when we need to perform ANOVA calculations on multiple sets of features in order to generate a list of p-values. A traditional approach is to use the base function aov and extract the p-value from it. This can be slow when analyzing thousands of features. An alternative is to use the multtest library …

Read More

Here’s an example of how to embed subtables in a datatable. It’s an enhancement of the child rows example found at https://rstudio.github.io/DT/002-rowdetails.html and works by storing JSON versions of subtables as a column in the main table dataframe. The current release of the DT library converts the JSON format into …

Read More

About 2 years ago my beloved Asus G60VX died. I loved that laptop. We'd been through a lot together and enjoyed five wonderful years. After a brief mourn I had a long think about the future and decided to move to a Chromebook. Precisely an Acer C270. I've moved from 15"x10" (16" screen), 7.3lbs to …

Read More

A common task when working with data is split-apply-combine – splitting a dataset up, applying a function and then recombining the results. It’s often performed in creating summary tables around a factor. First let’s define the data to be used. I'm interested in a scenario where I have nS samples at nC concentrations …

Read More