Skip to content

Graphic Design (Posters Figures Presentations)

Jump to bottom
Finlay Maguire edited this page May 20, 2021 · 7 revisions

Graphic Design

When it comes to designing posters, figures, presentations, or any other visual communication piece there are some great simple guidelines out there.

Generally:

  • 3 laws of effective visual communication: have a clear purpose, show the data clearly, and make the message obvious.

  • Focus on minimising text and simplifying graphics to distill the core message.

  • Think about where your eyes are drawn when looking at a graphic and make sure that's where you want your audiences eyes to go (in presentations you have the extra tools of overlay arrows/boxes and gradual reveals to do this)

  • Try and use the most appropriate graph for your data and what you are trying to show (and remove extraneous elements).

Here is a great cheatsheet of graphic design principles

Technical Tips

  • For posters and graphics I strongly recommend using a SVG based system e.g., inkscape although for paid tools adobe illustrator/indesign are most polished professional level tools. Powerpoint or equivalent (libreoffice/keynote/slides) are also used heavily in the life sciences for poster and infographic creation but if you have time to get over the learning curve I do recommend a more dedicated tool. LaTeX templates can also be used ideally via overleaf or similar and tikz in LaTeX is a powerful, if less user-friendly, way to create flowcharts/graphics programmatically.

  • For infographics you can find CC-0 (free to use without attribution - buts it is always nice if you can!) SVG icons etc available at svgrepo and there are also paid services like infographia. While far less nice, I also keep a repo of explanatory figures that I have made that might be usable as starting points for folks.

  • Colours: try and use a consistent set of colours across your poster/talk/figures/graphics. Colour theory for design has tons of resources out there but try to use a palette appropriate for your visualisation e.g., clearly separated colours for qualitative categories and a spectrum for quantitative. Always try to use colourblind friendly palettes if at all possible. colorbrewer is a handy resource for building palettes.

  • In terms of graphs/plots: highly recommend either creating or using in-built styles and palettes for the context the plot will be used in. For example if you are creating matplotlib-based plots, you can install seaborn (built on-top of matplotlib) and use its configs to automatically style your plots (even if they aren't created by seaborn as long as they use matplotlib). However, there are similar options in ggplot2 and other plotting libraries. Key configs to tweak or change depending on context:

  1. Set the context to poster or talk and it will increase the font size/ratios/line depths for any matplotlib based plots you make in the script/notebook to be more readable on a poster: documentation

  2. Set the palette being used to something like colorblind or an appropriate set from colorbrewer to avoid pitfalls with the plot colours: documentation

  3. Set a plot style (and use it consistently), I personally like whitegrid or ticks: documentation

  4. Clear figure labels/legends with appropriate level of detail for venue!

  5. Exporting figures: save and version control individual figures/graphics/posters as SVGs so they can be resized and tweaked later by hand if you need. In matplotlib etc this can be done via plt.savefig($name.svg). This way you don't have to worry about resolution etc. Otherwise try to ensure at least 300 DPI and appropriate cropping (again for matplotlib based plotting I recommend plt.savefig($name.png, dpi=300, bbox_inches='tight')).

Inkscape Specific Tips

There are a lot of guides/blogs/youtube videos out there for most things in inkscape but here are some key things to make life easier for yourself:

  • Check the overview of the interface explained in the documentation

  • Set the canvas size to the appropriate size for you use e.g., the poster dimensions using file -> document properties

  • Align and Distribute Menu is your friend for making things look neat.

  • Grouping and ungrouping items lets you build more complex units and/or break down and tweak individual graphical elements in exported graphs

  • Stroke and fill are how you control colour of objects and how thick the border line ("stroke") is around them, try to have an equivalent stroke size for all objects for visual consistency. You can also add a custom palette here such as the one you designed/chose using colorbrewer and guidance above.

  • Clipping and masking is how you can crop part of an imported picture or graphic in inkscape (although if you want to do more complex manipulation to image files then

  • Make sure any non-vector you import (e.g., images of graphs or something else; although please don't import an image of a table as often they are usually too low resolution and get warped) is appropriate resolution for your venue (e.g., DPI > 300 in most cases).

  • If exporting the whole poster/slide/graphic from SVG to something other than the usual PDF for talks/posters etc such as PNG then make sure it includes a background or it may display oddly. This can be set in file -> document properties -> background colour

Clone this wiki locally