Visit the Micm website for the latest workshops or simply sign up to our newsletter [ HERE ]
MiCM Workshops Series
As part of the McGill initiative in computational medicine (MiCM) Workshop Series,
This fall we will be offering another workshop.
Data Visualization in R: Essential and Optimization
An overview of essential plots and how to generate them using R for the exploration and presentation of data, tailored for biological datasets.
This workshop will also feature a hands-on, interactive introduction to the principles of good ﬁgure design, data visualization and ﬁgure optimization.
October 7, 2019
3506 Rue University, Montréal
QC H3A 2A7
9am - 4pm
Fee: $20.00 (space is limited)
To register: CLOSED
Download Poster micm_workshop_ad_october_7_2019.pdf
MiCM Workshops Series
Two workshops will be offered on behalf of the McGill initiative in computational medicine (MiCM) to those who want to learn visualization and analysis tools. For each workshop, attendees will receive, additional material prior to the workshop. The workshop content will be focused on exercises and practical computing.
1. July 11 2019, 9am-3pm Data visualization (ggplot)
2. Aug 13 2019, 9am-4pm R beyond the basics
McGill University Arts Building, room 150
853 Rue Sherbrooke Ouest
Montréal, QC H3A 2A7
Questions can be directed
to info-micm [at] mcgill.ca
Data visualization (ggplot) Overview
Data visualization (DV) is a form of visual communication used to present data, and is important across many disciplines. This course will cover the core principles of good graphics and best practices, how to choose a visualization approach depending on the type of data and the desired message, and demonstrations of how DV can yield insights into data otherwise lost or buried by statistics alone. Participants will also learn how to visualize data and implement basic graphics in R, and generate their own graphics to visualize data. Source code for demonstrated charts will be made available to workshop participants.
The focus of this course is on DV principles that translate to any tool or programming language. Although this workshop will discuss the implementation of common graphics in R as a demonstration, it not an R programming course and will not delve into the details of data manipulation techniques.
Principles of good graphics: Core principles of good DV, including the selection of the appropriate visualization strategy for the data and message, what makes an effective graphic, and what to avoid.
Types of visualization graphics: Survey various DV approaches for different types of data, from the basics to more specialized presentations (such as bar graphs andvariants, scatter plots, line charts, box plots, heatmaps, colourmaps, dendograms, and networks).
Visualizing data in R: How to visualize data and generate various graphic types in R. We will explore both methods that use build-in R functions and external libraries for visualization (such as ggplot2, Leaflet, RGL, Lattice, highcharter).
Check out what you missed [ photos ]
R Programming Beyond the Basics Overview
Statisticians are seeing large-scale and/or high-dimensional datasets more and more often. Computational efficiency is therefore essential. To achieve this, clean and succinct coding is often necessary although not sufficient. In this tutorial, we will start with efficient coding then transition to efficient computing.
First, we will implement different coding options to perform simple tasks such as calculating the mean gene expression levels of all different subgroups of a dataset stratified by age, gender, etc. We will learn how to write our own functions, or use powerful R functions including aggregate(), by() and the apply() family of functions to achieve such tasks, with short and clean codes. Through the parallel version of the apply() family in the ‘parallel’ package, the tutorial will transition to efficient computing. Various topics on computational efficiency will be covered, including the logistics, parallel computing, and linkage to other programming languages such as C++ (package ‘Rcpp’) and Fortran 90.