using visual aids effectively

Visual Aids

Phil Ewels

Using visual aids effectively

phil.ewels@scilifelab.se

What is a visual aid?

Figures Talk slides Posters

What is a

Figures Talk slidesPosters

• Helps your audience understand

• Often simple

• Interesting » Humour / Different / Interactive

good visual aid?

What is a

Figures Talk slidesPosters

good visual aid?

• Target audience / Content focus

• Good impressions / Visual aids

• Data visualisation / Workshop

Planning• Target audience

» One size does not fit all

» Be sympathetic - put yourselves in their shoes

» Aim for the lowest common denominator

• Content planning » Plan from the top down, not from the details up

» Write down an outline before you start

» Think about an “elevator pitch”

Elevator Pitch

distillation of idea

all work

core concept

Elevator Pitch

distillation of idea

all work

core concept

• Paper abstracts

• Graphical ToCs

• Talk poster / focus

• Coffee chat

Best / Worst Examples

Phil Ewels - Challenging samples for NGS / 20

Sample Setup

9

SciLifeLab ID Library Prep Starting amount Sequenced Reads

P1102_101 Manual 1000 ng 38,995,594

P1102_102 Manual 1000 ng 37,663,274

P1102_103 Manual 1000 ng 39,666,722

P1102_104 Manual 500 ng 35,332,272

P1102_105 Manual 200 ng 40,568,034

P1102_106 Manual 50 ng 47,044,650

P2011_1005 NeoPrep Run 1 25 ng 93,316,971

P2011_1006 NeoPrep Run 1 25 ng 115,648,988

P2011_1007 NeoPrep Run 1 25 ng 118,489,187

P2013_1004 NeoPrep Run 2 25 ng 72,128,476

P2013_1005 NeoPrep Run 2 25 ng 62,774,142

Phil Ewels - Challenging samples for NGS / 20

Sample Setup

10

Sample Library Prep Starting amount (ng) Sequenced Reads (M)

1 Manual

2 Manual

3 Manual

4 Manual

5 Manual

6 Manual

7 NeoPrep Run 1

8 NeoPrep Run 1

9 NeoPrep Run 1

10 NeoPrep Run 2

11 NeoPrep Run 2 25

25

25

25

25

50

200

500

1000

1000

1000

0 30 60 90 120

Visual Design

Introduction• Don’tunderes-matetheimpactofyourfirstfewslides

• Fontsandvisualpresenta-onimmediatelysetthetoneforyouraudience

• Anchoryourworkinthecontextofyouraudience’swork

• Goslow-everyonewillthankyouforit• Thisincludesnotusingtoomuchcontent• Trynottoreadeverybulletpointfromthescreen-talkaroundyourslidesinstead

• Don’tputallofyourbulletsupatonce,theaudiencewillreadtheminsteadoflisteningtoyou

• Nowistheperfect-metouseavisualaid

NO

Introduction• Don’t underestimate the impact of your first few slides

• Fonts and visual presentation immediately set the tone for your audience

• Anchor your work in the context of your audience’s work

• Go slow - everyone will thank you for it• This includes not using too much content• Try not to read every bullet point from the screen -

talk around your slides instead• Don’t put all of your bullets up at once, the

audience will read them instead of listening to you• Now is the perfect time to use a visual aid

Visual Design• Visual design is important

• Visual design is easy » Clear message

» Focussed

» Easy to read and interpret

» Honest and true reflection of the data

• Fonts. Colours. Layout.

Fonts

Body text

Figures

Presentations

Sans-serif Fonts

WARNINGserif fonts WARNING

serif fonts

Fonts• Pick a font and stick to it

• Avoid MS defaults » https://www.google.com/fonts

• Make use of font weights

Open sans Lato

Roboto (Arial)

Bold Medium Regular

Light Thin

Hairline

CambriaCalibri

Every time you use Comic Sans, Faye will punch this adorable little bunny.

comic sans criminal.com

Colour Palettes

Google Material Design Guidelines

Set of guidelines about design

Aimed for app developers

Includes some nice colour palettes

Lots of good stuff about design theory

https://www.google.com/design/spec/style/color.html

https://coolors.co

Data Visualisation

Choosing a plot• What type of graph best represents the

argument that you’re trying to make

• Which data are necessary

Relationship

Comparison

Composition

Distribution

What are you trying to show?

Distribution

Relationship / Comparison / Composition / Distribution

What are you trying to show?

Relationship

Relationship / Comparison / Composition / Distribution

What are you trying to show?

Composition

Relationship / Comparison / Composition / Distribution

What are you trying to show?

Comparison

Relationship / Comparison / Composition / Distribution

What are you trying to show?

What are you trying to show?

What are you trying to show?

• Distributions are not relevant

• Only the median value is needed

• Trends difficult to compare

• Easier comparison

Making Comparisons• Fast, accurate

judgement » Length

» Slope

• Medium judgement » Colour

» Patterns (eg. grouping)

• Slow, inaccurate judgement

» Angle

» Area

» Text

Making Comparisons• Length & Slope

» Bar / column charts » Box plots » Line graphs

• Colour » Heatmaps

• Patterns » Scatter plots

• Angle » Pie charts

• Area » Venn diagrams

• Text » Tables

» Plot labels

Making Comparisons

1234567

Making Comparisons1234567

0

0.5

1

1.5

2

1 2 3 4 5 6 7

Making Comparisons1234567

• Angles are bad for comparison

• Legend is disassociated from plot

• Requires colour link for series identification

0

0.5

1

1.5

2

1 2 3 4 5 6 7

Making Comparisons

Making Comparisons

Keep it simple

Keep it simple

0.0

0.5

1.0

1.5

2.0

1 2 3 4 5 6 7Series1 Series2

• Use of 3D was arbitrary

• No need for colour and texture

Aspect Ratio• Elements that need accurate aspect ratios:

» Images

» Text

» Anything circular

» Axes with comparable units

Aspect Ratio

Aspect Ratio

Aspect Ratio

-6

-4

-2

0

2

4

6

8

10

12

14

-6 -4 -2 0 2 4 6 8 10 12 14

log1

0 FP

KM

log10 FPKM

Aspect Ratio

-6

-4

-2

0

2

4

6

8

10

12

14

-6 -4 -2 0 2 4 6 8 10 12 14

log1

0 FP

KM

log10 FPKM

Features of a good plot

• Minimalistic

• Suitable plot type

• Big and clear

• Attractive (avoid defaults)

• The test - can you draw it from memory?

0

0.5

1

1.5

2

2.5

1 2 3 4 5 6 70.0

0.5

1.0

1.5

2.0

2.5

1 2 3 4 5 6 7

Using a graph

• Hold audience focus » Talk through your data

» Don’t show everything at once

• Use layering for complex plots » Progressively add data

Colour

Colour• Colour can be used to:

» Highlight specific data

» Group categories of data

» Encode quantitative values

• The more selective you are with colour, the greater its effect

0.0

0.2

0.5

0.7

0.9

1 2 3 4 5 6

0.0

0.2

0.5

0.7

0.9

1 2 3 4 5 6

Colour Scales• Represent quantitative data

• Sequential

• Divergent

• Categorical

Colour Sensitivity

Colour Sensitivity

Humans do not perceive the spectrum evenly Spectrum is a bad colour scheme with which to encode data

Bad DefaultsJet

Luminescence

afmhot

Luminescence

YlOrRd

Luminescence

not even

not even

Scale types

Categorical Sequential

Scale types

Sequential (should be categorical)

Which category is this?

Colour Blindness• Common in Northern European men

• Colour schemes such as Magenta – Green designed to be colour blind friendly

22% chance at least one colour blind

3 NE male reviewers

Colour Blindness

Normal Vision Protanopia

Colour Blindness

Normal Vision Protanopia

Color Brewer

http://colorbrewer2.org

Sequential Divergent Categorical

Summary

Summary• Decide your key points early

» Build around target audience

• Show only what you need to » Bullets, not prose

» Suitable graphs

» Avoid defaults

• Use data as a visual aid

colorbrewer2.org

google.com/fontsflaticon.com

coolors.co

Adobe Illustrator

Inkscape (free)

CreditsCourse written by Phil Ewels. Some material developed whilst working at the Babraham Institute in Cambridge, UK. Now working at the National Genomics Infrastructure, part of the Science for Life Laboratory in Stockholm, Sweden.

Find more at http://phil.ewels.co.uk