Twenty — minutes — maybe — more.choose four wordsmore quotes

# space: beautiful

UCD Computational and Molecular Biology Symposium, Dublin, Ireland. 1-2 Dec 2016.

# visualization + design

Cover image accompanying our article on mouse vasculature development. Biology turns astrophysical. PNAS 1 May 2012; 109 (18) (zoom, PNAS)

# Creating the PNAS Cover

One of my goals in life, which I can now say has been accomplished, is to make biology look like astrophysics. Call it my love for the Torino Impact Hazard Scale.

Recently, I was given an opportunity to attend to this (admittedly vague) goal when Linda Chang from Aly Karsan's group approached me with some microscopy photos of mouse veins. I was asked to do "something" with these images for a cover submission to accompany the manuscript.

When people see my covers, sometimes they ask "How did you do that?" Ok, actually they never ask this. But being a scientist, I'm trained me to produce answers in anticipation of such questions. So, below, I show you how the image was constructed.

The image was published on the cover of PNAS (PNAS 1 May 2012; 109 (18))

## Tools

Photoshop CS5, Nik Color Efex Pro 4, Alien Skin Bokeh 2 and a cup of coffee from a Rancilio Silvia.

## source images

Below are a few of the images I had the option to work with. These are mouse embryonic blood vessels, with a carotid artery shown in the foreground with endothelial cells in green, vascular smooth muscle cells in red and the nuclei in blue.

Of course, as soon as I saw the images, I realized that there was very little that I needed to do to trigger the viewer's imagination. These photos were great!

Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)
Mouse carotid arteries. (zoom)

## memories of star trek

Immediately I thought of two episodes of Star Trek (original series): Doomsday Machine and the Immunity Syndrome, as well as of images from the Hubble Telescope.

Enterprise is about to be consumed by a horror tube: a planet killer! (The Doomsday Machine)
Enterprise heads into a giant amoeba. Who eats whom? I'll let you guess. (The Immunity Syndrome)
Orion nebula (M42) as seen by the Hubble telescope. (zoom)

I though it would be pretty easy to make the artery images look all-outer-spacey. They already looked it.

## centerpiece image

And then I saw the image below.

A particularly spectacular image of a mouse carotid artery. I'm thinking 10 on the Torino scale. (zoom)

## constructing the cover

### background

The background was created from the two images shown here. The second image was sampled three times, at different rotations.

Images used for background. (zoom)
Images used for background. (zoom)
Layer composition for background elements. (zoom)

The channel mixer was used to remove the green channel and leave only red and blue.

Background elements for PNAS cover image. (zoom)

### middle ground

The next layer was composed of what looked like ribbons of blue gas. This was created by sampling the oval shapes from the source images. Here the red channel was a great source for cloud shapes, and this was the only channel that was kept. The hue was shifted to blue and a curve adjustment was applied to increase the contrast.

First set of middle ground elements, before adjustments. (zoom)
First set of middle ground elements, after channel adjustments. (zoom)
Second set of middle ground elements. (zoom)
Layer composition for middle ground elements. (zoom)

When the foreground and middle ground elements were combined, the result was already 40 parsecs away.

Background and foreground elements for PNAS cover image. (zoom)

### foreground

The foreground was created from the spectacular comet-like image of a mouse artery. Very little had to be done to make this element look good. It already looked good.

I applied a little blur using Alien Skin's Bokeh 2 to narrow the apparent depth of field, masked out elements at the bottom of the image and removed some of the green channel. The entire blue channel was removed altogether (this gave the tail of the comet a mottled, flame-like appearance).

Foreground element, before adjustments. (zoom)
Foreground element, after channel adjustments. (zoom)
Layer composition for foreground element. (zoom)

### post processing

Initial composition of background, middle ground and foreground elements. (zoom)
40% localized application of Nik's Tonal Contrast (Color Efex 4 plugin) to increase structure in red channel. (zoom)
50% blend with Nik's Pro Contrast (Color Efex 4 plugin). (zoom)

And here we have the final image.

Final PNAS cover. Spacey! (zoom)
VIEW ALL

# Intuitive Design

Thu 03-11-2016

Appeal to intuition when designing with value judgments in mind.

Figure clarity and concision are improved when the selection of shapes and colors is grounded in the Gestalt principles, which describe how we visually perceive and organize information.

One of the Gestalt principles tells us that the magenta and green shapes will be perceived as as two groups, overriding the fact that the shapes within the group might be different. What the principle does not tell us is how the reader is likely to value each group. (read)

The Gestalt principles are value free. For example, they tell us how we group objects but do not speak to any meaning that we might intuitively infer from visual characteristics.

Nature Methods Points of View column: Intuitive Design. (read)

This month, we discuss how appealing to such intuitions—related to shapes, colors and spatial orientation— can help us add information to a figure as well as anticipate and encourage useful interpretations.

Krzywinski, M. (2016) Points of View: Intuitive Design. Nature Methods 13:895.

# Regularization

Fri 04-11-2016

Constraining the magnitude of parameters of a model can control its complexity.

This month we continue our discussion about model selection and evaluation and address how to choose a model that avoids both overfitting and underfitting.

Ideally, we want to avoid having either an underfitted model, which is usually a poor fit to the training data, or an overfitted model, which is a good fit to the training data but not to new data.

Nature Methods Points of Significance column: Regularization (read)

Regularization is a process that penalizes the magnitude of model parameters. This is done by not only minimizing the SSE, $\mathrm{SSE} = \sum_i (y_i - \hat{y}_i)^2$, as is done normally in a fit, but adding to this minimized quantity the sum of the mode's squared parameters, $\mathrm{SSE} + \lambda \sum_i \hat{\beta}^2_i$.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Regularization. Nature Methods 13:803-804.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Model Selection and Overfitting. Nature Methods 13:703-704.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Classifier evaluation. Nature Methods 13:603-604.

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

# Model Selection and Overfitting

Fri 04-11-2016

With four parameters I can fit an elephant and with five I can make him wiggle his trunk. —John von Neumann.

By increasing the complexity of a model, it is easy to make it fit to data perfectly. Does this mean that the model is perfectly suitable? No.

When a model has a relatively large number of parameters, it is likely to be influenced by the noise in the data, which varies across observations, as much as any underlying trend, which remains the same. Such a model is overfitted—it matches training data well but does not generalize to new observations.

Nature Methods Points of Significance column: Model Selection and Overfitting (read)

We discuss the use of training, validation and testing data sets and how they can be used, with methods such as cross-validation, to avoid overfitting.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Model Selection and Overfitting. Nature Methods 13:703-704.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Classifier evaluation. Nature Methods 13:603-604.

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

# Classifier Evaluation

Tue 13-09-2016

It is important to understand both what a classification metric expresses and what it hides.

We examine various metrics use to assess the performance of a classifier. We show that a single metric is insufficient to capture performance—for any metric, a variety of scenarios yield the same value.

Nature Methods Points of Significance column: Classifier Evaluation (read)

We also discuss ROC and AUC curves and how their interpretation changes based on class balance.

Lever, J., Krzywinski, M. & Altman, N. (2016) Points of Significance: Classifier evaluation. Nature Methods 13:603-604.