π Daylatest newsbuy art
Embrace me, surround me as the rush comes.Motorcycledrift deeper into the soundmore quotes
science + communication
Complaining about ineffective graphics is easy — fixing them takes time and love. After receiving a migrane-inducing graphic, I set out on a mission to tell the story of the data — obesity patterns around the world.
Visit the Poster Hospital to see redesigns of real-world posters and learn practical design guidelines for scientific posters and layouts on a large canvas.
The Nature Methods Points of View column column offers practical advice in design and data presentation for the busy scientist.

Effective graphical abstract design for science communication

Guidelines to get you started and keep you going

Martin Krzywinski / Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Guidelines for summarizing your research story with a graphical abstract—sober design, typography and data visualization tips all in one place with a minimum of fuss (v1.2 12 Nov 2020). (Download PDF template)


The graphical abstract guidelines build on my design guidelines for posters. Because their focus is on telling the essence of a story in a small space, I see them as much less negotiable than guidelines for posters, where the large canvas allows for more options.

See how the guidelines can be applied to graphical abstracts in the wild by visiting the graphical abstract hospital.

The Graphical abstract mindset

# 1
Pure story clearly told

The purpose of the graphical abstract is to quickly present the key findings of a paper in a way that closely matches the narrative of the paper. The graphical abstract should place all central terms in context and emphasize key relationships.


A graphical abstract is Figure 0.


Everything you learn about the design of graphical abstracts applies to the design of figures.


Everything you learn about the design of figures applies to the design of posters.

Identifying what to put in the graphical abstract cannot be answered by design alone. Design can help place and shape what you have chosen, but it cannot tell you what to chose — that's up to the science.

Graphical abstracts are typically small square bitmaps. Though individual journal requirements vary, this constraint on the size of the canvas places tremendous limits on the design and imposes more strict requirements than when designing for a larger space.


The graphical abstract is a tiny space that offers no quarter. A bad design decision cannot be mitigated by a good design decision downstream. There is no room for "style".


The graphical abstract is a visual elevator pitch. There is no room for visual garnish or embelishment of any kind.


The function of a graphical abstract comes from its form, which cues the nature of relationships in its story.

Many abstracts struggle to fit the path of the story into a small space. Individual scenes meander down the image, often snaking, as elements struggle for space and attention. Make it easy for the eye and guide it along simple paths.

# 2

The graphical abstract should never be designed with attracting the eye as the first priority. Rather, it's your first opportunity for engaging the reader's pre-attentive processing and to lay out a map of the story. Every part of the abstract should correspond to the beginning of important story arcs in the paper.


Strict control of color.


No gradients. There, I said it.


No elaborate three-dimensional representations. While well-drawn ones can be attractive, many distract the eye with too much detail and draw attention away from the story path.

Your paper is a mansion and the graphical abstract is a studio apartment — you cannot bring all the furniture. Too often, "style" is that credenza you've been lugging around everywhere. It's time to let it go.

# 3

Most graphical abstracts have poor top-level organization. In many, the wrong things are grouped and categories, such as experimental conditions, are not clearly demarcated. Opportunities for maintaining continuity between scenes of the story are often missed.


Understand and practise good Gestalt grouping principles. In a small space, color will group more powerfully than position and much more powerfully than shape.


Organize content in a way that reflects spatial or semantic hierarchy by making use of Gestalt figure-ground principles. These are very helpful in reinforcing inside vs. outside (e.g. cytosol vs. nucleus).


Tables are dead. Long live the table.

Typical graphical tropes, such as those used in figures and presentations, have no place in graphical abstracts. Everything must be distilled down to its core essence — this part is hard and requires that you pay attention to how graphical elements relate in shape and position and whether these relationships are unambiguous and speak to your story with specificity.

Many abstracts do not balance visual weight: they have too much detail in shapes and symbols and have too much variation in text size and line thickness.

In most abstracts, key aspects of the story are lost among too much variation due to loose design.


For every pair of things on the abstract, ask yourself: (1) can their graphical relationship be improved to better match their actual relationship? (2) does their graphical relationship imply something that isn't important or true? (3) am I repeating myself?


Are things of equal importance drawn with equal emphasis? How many different sizes of type are being used?


The abstract's grid layout should present a strong tabular feel. Can you draw a horizontal or vertical line across the abstract to divide its elements along thematic or categorical boundaries?

# 4
Leading the eye

Stick to straight paths for the eye: left-to-right, top-to-bottom, or top-left to bottom-right. Avoid requiring the reader to backtrack or turn sharply.


Align ruthlessly. In a small space even tiny misalignments become obvious and a barrier to assessing relationship between elements.


Encapsulate your story in a strong and obvious grid layout (1×2, 2×2, ...) and maintain symmetry in margins, padding and spacing between elements.


Use open form (elements extending beyond the image) to imply that the scene is a vignette of a larger scene or more complex process.


# 5
More color isn't the solution

The success of a graphical abstrat rests on how well it is organized and how well it represents important relationships. Color can be used to maintain a theme (e.g. health vs disease) but this effect is diluted by other colors.


Do not create too many color groups. Instead, group using space and shape.


Avoid strongly coloring elements that are drawn for context or are of secondary importance.


Avoid background color, which generally reduces contrast. If you cannot overcome your desire for a background gradient, make it as subtle as possible.


# 6
Tightly control font size and color

Well-chosen type placement and size goes a long way in alleviating tightness of a small space.


Have a modest type size ladder with no more than 3 sizes — for example, 4, 6, 8 pt.


Generally, do not apply color to text unless you're cuing the continuation of a theme. But do so carefully.


Use light and dark grey for text to indicate a hiearchy of importance.

# 7
Do not rely on a variety of type faces (bold, italic) or styling

On a large canvas, there is room for a fully type hierarchy (bold, small caps, italics, etc). In a small space, there isn't. Always evaluate the legibility of text on the final bitmap.


Do not outline or add styling to text, such as drop shadows or glow.


Apply bold and italics gently and do not combine them — too much variation in type will overwhelm the space.


Small italic labels make for great annotations.

Technical Considerations

# 8
Design with physical size in mind

Although the graphical abstract itself may never be printed, it's extremely helpful to design it to a specific physical size. You'll be able to standardize your design process for graphical abstracts, figures and posters.


Design on a 4" × 4" artboard and export at 300 dpi. Scale artboard size, not output resolution, to match journal requirements.


Use point units for sizing text in the range 6–10 pt. Limit yourself to Arial or one of the Helveticas. Avoid angled or vertical text.


Draw lines no thinner than 0.5 pt. Standardize line and arrow length and angle.

news + thoughts

How Analyzing Cosmic Nothing Might Explain Everything

Thu 18-01-2024

Huge empty areas of the universe called voids could help solve the greatest mysteries in the cosmos.

My graphic accompanying How Analyzing Cosmic Nothing Might Explain Everything in the January 2024 issue of Scientific American depicts the entire Universe in a two-page spread — full of nothing.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
How Analyzing Cosmic Nothing Might Explain Everything. Text by Michael Lemonick (editor), art direction by Jen Christiansen (Senior Graphics Editor), source: SDSS

The graphic uses the latest data from SDSS 12 and is an update to my Superclusters and Voids poster.

Michael Lemonick (editor) explains on the graphic:

“Regions of relatively empty space called cosmic voids are everywhere in the universe, and scientists believe studying their size, shape and spread across the cosmos could help them understand dark matter, dark energy and other big mysteries.

To use voids in this way, astronomers must map these regions in detail—a project that is just beginning.

Shown here are voids discovered by the Sloan Digital Sky Survey (SDSS), along with a selection of 16 previously named voids. Scientists expect voids to be evenly distributed throughout space—the lack of voids in some regions on the globe simply reflects SDSS’s sky coverage.”


Sofia Contarini, Alice Pisani, Nico Hamaus, Federico Marulli Lauro Moscardini & Marco Baldi (2023) Cosmological Constraints from the BOSS DR12 Void Size Function Astrophysical Journal 953:46.

Nico Hamaus, Alice Pisani, Jin-Ah Choi, Guilhem Lavaux, Benjamin D. Wandelt & Jochen Weller (2020) Journal of Cosmology and Astroparticle Physics 2020:023.

Sloan Digital Sky Survey Data Release 12

constellation figures

Alan MacRobert (Sky & Telescope), Paulina Rowicka/Martin Krzywinski (revisions & Microscopium)


Hoffleit & Warren Jr. (1991) The Bright Star Catalog, 5th Revised Edition (Preliminary Version).


H0 = 67.4 km/(Mpc·s), Ωm = 0.315, Ωv = 0.685. Planck collaboration Planck 2018 results. VI. Cosmological parameters (2018).

Error in predictor variables

Tue 02-01-2024

It is the mark of an educated mind to rest satisfied with the degree of precision that the nature of the subject admits and not to seek exactness where only an approximation is possible. —Aristotle

In regression, the predictors are (typically) assumed to have known values that are measured without error.

Practically, however, predictors are often measured with error. This has a profound (but predictable) effect on the estimates of relationships among variables – the so-called “error in variables” problem.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Error in predictor variables. (read)

Error in measuring the predictors is often ignored. In this column, we discuss when ignoring this error is harmless and when it can lead to large bias that can leads us to miss important effects.

Altman, N. & Krzywinski, M. (2024) Points of significance: Error in predictor variables. Nat. Methods 20.

Background reading

Altman, N. & Krzywinski, M. (2015) Points of significance: Simple linear regression. Nat. Methods 12:999–1000.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of significance: Logistic regression. Nat. Methods 13:541–542 (2016).

Das, K., Krzywinski, M. & Altman, N. (2019) Points of significance: Quantile regression. Nat. Methods 16:451–452.

Convolutional neural networks

Tue 02-01-2024

Nature uses only the longest threads to weave her patterns, so that each small piece of her fabric reveals the organization of the entire tapestry. – Richard Feynman

Following up on our Neural network primer column, this month we explore a different kind of network architecture: a convolutional network.

The convolutional network replaces the hidden layer of a fully connected network (FCN) with one or more filters (a kind of neuron that looks at the input within a narrow window).

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Convolutional neural networks. (read)

Even through convolutional networks have far fewer neurons that an FCN, they can perform substantially better for certain kinds of problems, such as sequence motif detection.

Derry, A., Krzywinski, M & Altman, N. (2023) Points of significance: Convolutional neural networks. Nature Methods 20:1269–1270.

Background reading

Derry, A., Krzywinski, M. & Altman, N. (2023) Points of significance: Neural network primer. Nature Methods 20:165–167.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of significance: Logistic regression. Nature Methods 13:541–542.

Neural network primer

Tue 10-01-2023

Nature is often hidden, sometimes overcome, seldom extinguished. —Francis Bacon

In the first of a series of columns about neural networks, we introduce them with an intuitive approach that draws from our discussion about logistic regression.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Neural network primer. (read)

Simple neural networks are just a chain of linear regressions. And, although neural network models can get very complicated, their essence can be understood in terms of relatively basic principles.

We show how neural network components (neurons) can be arranged in the network and discuss the ideas of hidden layers. Using a simple data set we show how even a 3-neuron neural network can already model relatively complicated data patterns.

Derry, A., Krzywinski, M & Altman, N. (2023) Points of significance: Neural network primer. Nature Methods 20:165–167.

Background reading

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of significance: Logistic regression. Nature Methods 13:541–542.

Cell Genomics cover

Mon 16-01-2023

Our cover on the 11 January 2023 Cell Genomics issue depicts the process of determining the parent-of-origin using differential methylation of alleles at imprinted regions (iDMRs) is imagined as a circuit.

Designed in collaboration with with Carlos Urzua.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Our Cell Genomics cover depicts parent-of-origin assignment as a circuit (volume 3, issue 1, 11 January 2023). (more)

Akbari, V. et al. Parent-of-origin detection and chromosome-scale haplotyping using long-read DNA methylation sequencing and Strand-seq (2023) Cell Genomics 3(1).

Browse my gallery of cover designs.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
A catalogue of my journal and magazine cover designs. (more)
Martin Krzywinski | contact | Canada's Michael Smith Genome Sciences CentreBC Cancer Research CenterBC CancerPHSA
Google whack “vicissitudinal corporealization”
{ }