Hands-on Biological Data Visualization with ggplot2 & R

Biological data visualization is an important aspect of bioinformatics which involves the graphical representation of unstructured or structured data to display information hidden in the plots/graphs. The ggplot2 package of R provides various functions to create different graphs and allows modification, annotation, and much more. Learning biological data visualization can not only help you in creating beautiful data visualization but also in making impactful decisions during your research based on data visualizations along with publishable figures for your research papers.

BioCode is offering an advanced biological data visualization with ggplot2 & R course in which you’ll learn how to write customized scripts to generate publication-ready and high-quality graphical plots of biological datasets to visualize, analyze and compare the datasets in a more insightful way, such as volcano plots, heatmap, dot plots, frequency plots, scatter plots, histograms, and bar charts with customized labeling. Even if you don’t know programming or R language at all, in this workshop you’ll learn R language from the basics to the advanced level and then apply your script-writing skills to data visualizations.

Whether you’re a beginner in bioinformatics, an experimental biologist, or a bioinformatics researcher, our course will help you greatly. You’ll be practicing on a case study of human mitochondrial proteome and genome along with a real-world gene expression dataset. Therefore, you’ll not only learn data visualization with ggplot2 in R, but you’ll also do real-world biological data visualization in this biological data visualization with ggplot2 in R workshop.

This course will include the following section:

Section 1: Data Visualization Using ggplot2

Description: This section will focus on making sure that the students learn how biological data is visualized using the ggplot2 package in R language.

Learning Outcomes:  Upon completion of this section, students will be able to:

1. Explain ggplot2 and its Use in Biological Data Representation.
2. Describe Key Components in ggplot2.
3. Visualize Human Mitochondrial Proteome.
4. Facet the Human Chromosome Dataset.
5. Smooth Out the Biological Data.
6. Create Box Plot for Human Mitochondrial Proteome.
7. Create Histograms for Human Mitochondrial Pattern Finding.
8. Create Frequency Plots for Human Mitochondrial Information Frequency Mining.
9. Create Bar Charts for Human Mitochondrial Knowledge Mining.
10. Scale and Limit Data Visualization.
11. Change Labels and Finalize Visualization.
12. Visualize a Phylogenetic Tree.
13. Save Visualizations in High Resolution.
14. Find Differentially Expressed Genes With Volcano Plot Visualization.

Make figures that carry your argument, with ggplot2

A figure is how most people will encounter your result. This course teaches R and ggplot2 from the beginning, across five segments, using biological data throughout — so what you practise on is the kind of data you actually have.

The five segments

• R and installation — getting a working environment, and orienting yourself in RStudio.
• Functions and variables — the language basics, taught only as far as plotting requires.
• Packages and data types — how R stores data, which is where most beginners get stuck.
• Control flow and data pre-processing — reshaping biological data into the form a plot needs. In practice this is most of the work.
• Visualisation with ggplot2 — building plots layer by layer, and controlling every element of their appearance.

Why ggplot2 specifically

ggplot2 implements a grammar of graphics: you declare what maps to what — this variable to the x axis, that one to colour — and the library composes the figure. Once that idea lands, producing an unfamiliar plot type stops being a search for the right function and becomes a matter of describing what you want. It is also the standard in genomics, so most published R figure code you encounter will be ggplot2.

The part people underestimate

Most time spent “making a figure” is spent reshaping data, not plotting it. Counts in the wrong orientation, conditions split across columns that should be rows — this course spends a full segment on that, because it is the actual obstacle.

What you can do afterwards

Take a biological dataset and produce publication-quality figures: expression heatmaps, volcano plots, PCA plots, boxplots with proper statistical annotation — and control colour, scale, labelling and layout well enough to meet a journal’s requirements.

Who it suits

Researchers who currently make figures in Excel or GraphPad and want reproducible ones, students preparing a thesis, and anyone whose plots are fine but take too long to produce. No prior R experience is assumed.