‘Points of View’ is a monthly column published by Nature Methods that deals with the fundamental aspects of visual presentation applicable to anyone who works with visual representation of data. Each month I focus on a particular aspect of data presentation or visualization and provide easy-to-apply tips on how to create effective presentations.
Below is a short excerpt from past columns. The current issue of Nature Methods can be found here.
Visualizing biological data — Data visualization is increasingly important, but it requires clear objectives and improved implementation.
by Bang Wong
December 2012
Researchers today have access to an unprecedented amount of data. The challenge is to benefit from this abundance without being overwhelmed. Data visualization for efficient exploration and effective communication is integral to scientific progress. For visualization to continue to be an important tool for discovery, its practitioners need to be present as members of research teams.
Pencil and paper — A unique set of tools facilitate thinking and hypothesis generation.
by Bang Wong & Rikke Schmidt Kjærgaard
November 2012
Creating pictures is integral to scientific thinking. In the visualization process, putting pencil to paper is an essential act of inward reflection and outward expression. It is a constructive activity that makes our thinking specific and explicit. Compared to other constructive approaches such as writing or verbal explanations, visual representation places distinct demands on our reasoning skills by forcing us to contextualize our understanding spatially.
Power of the plane — Two-dimensional visualizations of multivariate data are most effective when combined.
by Nils Gehlenborg & Bang Wong
October 2012
High-dimensional data pose a significant analytical and representational challenge. One instinctual response has been to represent data in three-dimensional (3D) space in order to capture additional information1. Given the common medium utilized for science communication, great utility can be achieved by pushing the communicative power of the endless 2D planes that surround us in the form of pieces of paper, computer monitors and video projections.
Nature Methods Sept. 2012
by Nils Gehlenborg & Bang Wong
September 2012
When working with high-dimensional data, it may be tempting to choose a three-dimensional (3D) spatial visualization over a two-dimensional (2D) ‘flat’ representation because it allows us an additional data dimension. However, because quantitative, categorical and relational data are often not representing spatial relationships, plotting them in 3D space adds a level of visual complexity that often makes the data more difficult to understand. It therefore can be more effective to plot these data on a 2D plane and rely on nonspatial graphical encodings to represent additional dimensions.
Nature Methods Aug. 2012
by Nils Gehlenborg & Bang Wong
August 2012
Data can be classified in many ways. One useful method of classifying data for visualization is to distinguish between those with and without an inherent order. For example, a set of species (such as Escherichia coli, Drosophila melanogaster and Homo sapiens) has no intuitive ordering and is considered ‘categorical data’, whereas a list of gene expression values is ‘ordered data’ because we can sort them from lowest to highest. In a previous column, we described methods for color-coding categorical data (August 2010). Here we focus on creating color maps for quantitative data.
Nature Methods July 2012
by Cydney Nielsen & Bang Wong
July 2012
With a rapidly growing collection of genomes coming from such initiatives as the 1000 Genomes Project, the days of a single reference genome are numbered. Although the genomic sequence between any two human individuals differs only by about 0.1%, there are abundant structural and copy-number variations of different types and sizes. Effective visualization of these genomic variations is required to gain insight into the genetic basis of human health and disease. However, variation data pose new challenges to traditional genome visualization tools, which depend on linear layouts and have difficulty depicting large structural rearrangements.
Nature Methods June 2012
by Cydney Nielsen & Bang Wong
June 2012
Obtaining genome-scale data has never been easier. In addition to sequencing genomes, biologists now routinely profile epigenomes, transcriptomes and proteomes. There are exciting opportunities to better understand genome regulation by integrating diverse data types into unified views. Visualization facilitates data interpretation, but designing meaningful visual depictions of these data is a challenge.
Nature Methods May 2012
by Cydney Nielsen & Bang Wong
May 2012
Many genomics techniques produce measurements that have both a value and a position on a reference genome. The genome coordinate provides a natural ordering to these data values and is the organizing principle driving how we commonly display and navigate genomic data today. A popular plotting approach is to arrange the linear genome coordinate along the x axis and express the data value range on the y axis. This conventional representation is limiting. By using other organizational frameworks we can better extract the information of interest and make sense of its patterns.
Nature Methods Apr. 2012
by Nils Gehlenborg & Bang Wong
April 2012
Different data types have their own inherent structure that makes specific visualization techniques most fitting. For example, a matrix of gene expression values for given cell measurements can be highly informative when displayed as a heat map or parallel coordinate plot. The challenge is finding visualizations that will effectively combine data types. Many research studies depend on integrating data to comprehend underlying processes. Here we explore ways to merge data that are best represented as heat maps and node-link diagrams: two common but disparate graphing techniques.
Nature Methods Mar. 2012
by Nils Gehlenborg & Bang Wong
March 2012
Heat maps represent two-dimensional tables of numbers as shades of colors. This is a popular plotting technique in biology, used to depict gene expression and other multivariate data. The dense and intuitive display makes heat maps well-suited for presentation of high-throughput data. Hundreds of rows and columns can be displayed on a screen. Heat maps rely fundamentally on color encoding and on meaningful reordering of the rows and columns. When either of these components is compromised, the utility of the visualization suffers.
As of February 26, 2012 – #1 Most downloaded at Nature Methods.
Nature Methods Feb. 2012
by Nils Gehlenborg & Bang Wong
February 2012
Most biological phenomena arise from the complex interactions between the cell’s many constituents such as proteins, DNA, RNA and small molecules. The graphical representations of networks can be useful in exploring this complex web of interactions. Choosing a suitable network visualization based on the patterns one hopes to highlight can yield meaningful insights into data.
Nature Methods Jan. 2012
by Noam Shoresh & Bang Wong
January 2012
Data visualization can serve two distinct purposes: to communicate research findings and to guide the data-exploration process as the scientific story is unfolding. Each goal entails a different approach to data representation, but sound graphic design principles are important in both. This column is the first in a series that will focus on data-visualization techniques intended to support data exploration.
Nature Methods Dec. 2011
by Bang Wong
December 2011
The primary tenets of design are utility and function. Just as objects are intuitive to use when they are well-designed, thoughtfully conceived scientific figures, slides and posters can be easy to interpret and understand. Whereas industrial design focuses on things people use, graphic design is concerned with designs people read. The design process helps us develop a visual literacy to construct presentations that are appealing and convincing.
Nature Methods Nov. 2011
by Bang Wong
November 2011
In science communication, it is critical that visual information be interpreted efficiently and correctly. The discordance between components of an image that are most noticeable and those that are most relevant or important can compromise the effectiveness of a presentation. This discrepancy can cause viewers to mistakenly pay attention to regions of the image that are not relevant. Ultimately, the misdirected attention can negatively impact comprehension.
Nature Methods Oct. 2011
by Bang Wong
October 2011
Layout is the act of arranging text and images on the page according to an overall aesthetic scheme and for the purpose of clarifying a presentation. In graphic arts, it is the elephant in the room; layout underlies everything we do when we communicate visually. Well-structured content can guide readers through complex information, but when the material we present lacks order, it can confuse or, worse yet, agitate readers trying to make sense of the material.
Nature Methods Sept 2011
by Bang Wong
September 2011
Arrows are one of the most commonly used graphical devices in scientific figures. In the July 2011 issue of Nature Methods alone I counted nearly 300 instances of arrows; more than half of the figures contain them. Given the widespread use of arrows, it is worthwhile to take a closer look at this privileged class of diagrammatic form and how we might benefit from its use.
Nature Methods Aug 2011
by Bang Wong
August 2011
In the past two columns I have focused on making information accessible. I discussed ways to avoid color and shift color hues to make them discernible by individuals with color vision deficiencies. In this column I focus on ways to make information apparent by simplifying its presentation.
As of August 30, 2011 – #2 Most Emailed content at Nature Methods.
Nature Methods July 2011
by Bang Wong
July 2011
Last month I wrote about color blindness and ways to make information accessible to individuals with color vision deficiencies. I would like to continue by considering graphical alternatives to color that could improve the overall clarity and utility of data displays.
Nature Methods Jun 2011
by Bang Wong
June 2011
Since my first column on color coding1 appeared, we have received a number of e-mails asking us to highlight the issue of color blindness. One of those correspondences was published in the October 2010 issue. Here I offer guidelines to make graphics accessible to those with color vision deficiencies.
Nature Methods May 2011
by Bang Wong
May 2011
Our goal when writing research papers is to convey information as clearly as possible. In past columns I have suggested several graphic design techniques to improve the clarity of figures. In addition to refining data figures, including overview figures in a research paper provides a framework for readers to understand the experimental design and reported findings.
Nature Methods Apr 2011
by Bang Wong
April 2011
Typography is the art and technique of arranging type. Like a person’s speaking style and skill, the quality of our treatment of letters on a page can influence how people respond to our message. It is an essential act of encoding and interpretation, linking what we say to what people see.
Nature Methods Mar. 2011
by Bang Wong
March 2011
I will continue to demonstrate how judicious choice of graphical representations can improve visual communication. Here I will focus on data figures.
The power and primary purpose of graphs is to reveal connections in data. As opposed to tables, in which there is little visual association between individual values, graphs and charts depend on readers to form patterns. In reading graphs, we observe individual data points, keep each of them in memory and construct an image from the constituents. The entire process can be exceedingly fast and attest to the power of visual perception. Graphical encoding needs to support the detection and assembly process of reading graphs.
Nature Methods Feb 2011
by Bang Wong
February 2011
My goal over the next two months is to show concretely how scientific figures can benefit from design principles. I will review concepts from past columns by applying them to several published figures.
In the design of common objects, such as a door, when a handle is used many people will mistakenly pull even if the door is to be opened by pushing. When the handle is replaced with a flat plate, which affords pushing, people will know to push. When dealing with figures, we depend on visual cues. We want our figure’s layout to express its underlying meaning.
Nature Methods Jan 2011
by Bang Wong
January 2011
Negative space, also known as whitespace, refers to the unmarked areas of the page. Collectively, it is the margins and the gaps between text blocks and images. Whitespace is as much a part of a composition as the titles, words and pictures. The Swiss typographer Jan Tschichold calls whitespace ‘the lungs of a good design’. In addition to giving elements breathing room, judicious use of whitespace can dramatically improve the visual appeal and effectiveness of figures, posters and slides.
Nature Methods Dec 2010
by Bang Wong
December 2010
Our visual system attempts to structure what we see into patterns to make sense of information. The Gestalt principles describe different ways we organize visual data. Last month, we looked at four principles that incline us to group objects when they are made to look alike, are placed near one another, are connected by lines or are enclosed in a common space (1). This month, we will examine the principles of visual completion and continuity. These principles are useful in page layout work and when we compose figures and slides.
Nature Methods Nov 2010
by Bang Wong
November 2010
Gestalt principles of perception are theories proposed by German psychologists in the 1920s to explain how people organize visual information. Gestalt is a German word meaning shape or form. The principles describe the various ways we tend to visually assemble individual objects into groups or ‘unified wholes’. They are highly relevant to the design of charts and graphs as well as the reports that contain them.
Nature Methods Oct. 2011
by Bang Wong
October 2010
In last month’s column we explored ways to encode data that enhance ‘accuracy’ when readers decode information from graphs. This month, we will focus on salience as a way to differentiate graphical symbols and improve ‘speed’ when reading graphs.
Nature Methods Sept. 2011
by Bang Wong
September 2010
Data figures or graphs are essential to life-science communication. Using these tools authors encode information that readers later decode. It is imperative that graphs are interpreted correctly. Despite the importance and widespread use of graphs, we primarily rely on our intuition, common sense and precedent in published material when creating them—a largely unscientific approach.
Nature Methods Aug 2010
by Bang Wong
August 2010
Color can add dimensionality and richness to scientific communications. In figures, color is typically used to differentiate information into classes. The challenge is picking colors that are discriminable. A systematic approach to choosing colors can help us find a lineup effective for color coding.
As of October 7, 2010 – #1 Most downloaded and #4 Most emailed content at Nature Methods.
