This love loves love. It's a strange love, strange love.find a way to lovemore quotes

# science: beautiful

In Silico Flurries: Computing a world of snow. Scientific American. 23 December 2017

# Nature Methods: Points of View

Points of View column in Nature Methods. (Points of View)

## Guidelines for Effective Figures

Practical and concise advise on the visual presentation of data for researchers. One topic and one page at a time.

## Common Challenges in Figure Design

Andreas Dahlin runs a figure making course at Uppsala University. He was kind to share with me common questions and concerns that his students have when creating figures (emphasis is mine).

I face problems for using the tools in power point to make nice illustration figures, and in addition how one can enhance the resolution of the figures to print it in a high quality mode.

In my opinion, the most difficult thing is how to draw the good-looking pictures and design the structure of slide to make it simple and substantial in content.

I find it difficult to find the right software to draw pictures.

The most difficult thing for me, when I make a figure, is to arrange the parts of the figure in a way they look nice and understandable.

I think the most difficult part is creating the concept, how to make a figure easy and fast to understand but not lacking all essential parts.

Stepping outside of my own knowledge of what the picture presents and viewing it as someone who sees it for the first time. It's easy to assume that some things are self evident and not making them clear enough in the pictures.

Figures that not are plots can also be tricky to get to look nice.

Anytime you have to draw something in paint, gimp, or other image program it requires a lot of work to make it look even slightly better than crap.

The most difficult thing (in general) is to include as much information as possible and display it in a way that is easy to understand. Figures should be intuitive for the reader, which is sometimes difficult to achieve. There might also be technical difficulties in achieving what you've visualized.

I think the most difficult part for me is to highlight the main idea I would like to express.

For me the most difficult part is making 3-D figures. Also while making figures its hard to decide on the good colors to choose for the figure.

In my opinion, the most difficult part when making a figure is don't know which software we can use and how to use.

The most difficult part for me is to start it! Because I am so meticulous and I am a painter, then it is not so easy to make decision about my figures and which one is better and so on, then finally I give up and put just one figure which of course I don't like...

I think it is difficult to put together my ideas to something that is connected and makes it easier for the viewer to understand.

It is so easy to just get an image from internet. I don’t know what is ok to do. There seems to be different rules in different communities.

To come up with a figure that does not simplify the concept too much at the same time as it does not overwhelm the viewer. To get some ideas for this is the reason why I take the course. ;-)

To me, how to make it easy to understand is the difficult part.

I think it is to save it in the correct format: Raster or vector, png or jpg or pdf... especially if I want to make some changes in the future to the figure.

I think is to choose the most appropriate figure that really help to transmit the information we want. Then, how many words can be good enough for been part of the message. At the beginning I used to use too many.

Apart from the difficulty of making the figure clear and easy to understand, the biggest problem I'm having is the captions. How long and detailed description is appropriate, so it neither steals attention from the figure nor leaves out too much important information.

I think the most difficult part is to have high resolution image once we want to save it. My experience is when finish with drawing, the file size sometimes to large for high quality image and if we downgrade it, the image becomes bad.

The most difficult part when i making a figure is the software using part, I'm not good at computer so that part is annoying for me all the time.

I think the most difficult is to find out how to condensate many ideas in one picture without making it difficult to understand.

The most difficult part is the get the image to not look too amateurish that people focus on that instead of the message.

The most difficult part when doing a figure is to let it speak for itself, i.e. to not have long caption text.

To be able to depict all the desirable results on a single figure is sometimes not that easy. It becomes more critical when a figure is to be fitted within a certain size frame. An exact placing of a figure in some text editors often comes along with difficulties.

The most difficult part when making a figure is to make it simple and still be informative.

Depends a lot on the kind of figure, but generally it is to get clarity in the design, such that the idea is conceived easily. This requires some good outline (usually an iterative process).

The most difficult part to make a figure is the need to express abstract concepts into drawings.

The compromise between include detailed information and at the same time be readable (figures in articles)

To compress all information and ideas you have in your head into short and clear message.

I feel the difficulty in choosing a right resolution of the picture and the angle that could visualize all the details. And also choosing right test/label colour, size, font. Another difficulty for me is continuation from one slide to another.

I believe that my biggest problem would be making nice flux charts. Generally the ones I draw look too crude, it does not look beautiful. I have no concern about making an image that can represent an idea, but making a beautiful image makes it more pleasing to the eyes of the people who will read my work.

It is very difficult to make the figure delicate. I am still not get used to put all the small components together to integrate the figure by the vector software, instead of drawing it out directly.

I think the most difficult part is to make the image simple but yet informative.

I find it very difficult to make an original clarity picture in a particular format after dimensioning it according to the requirement.

Some times it is difficult to limit the size (Bytes) of the picture when going for high clarity remake.

Making the figure as informative as you want while keeping it simple enough to grasp quickly.

For me, the more difficult part is to create a figure that contains or tells all the information that I want to transmit, but keeping the figure simple, clean and not overloaded.

The most difficult for me is make it easily to be understood meanwhile containing the essential information.

The most difficult thing when developing a figure is ... to remove the bloat but keep the message. (Besides the very most difficult: finding out what I want to tell.)

For me the most difficult part is to choose colors with right contrast and to make it more attractive and catchy.

VIEW ALL

# Statistics vs Machine Learning

Tue 03-04-2018
We conclude our series on Machine Learning with a comparison of two approaches: classical statistical inference and machine learning. The boundary between them is subject to debate, but important generalizations can be made.

Inference creates a mathematical model of the datageneration process to formalize understanding or test a hypothesis about how the system behaves. Prediction aims at forecasting unobserved outcomes or future behavior. Typically we want to do both and know how biological processes work and what will happen next. Inference and ML are complementary in pointing us to biologically meaningful conclusions.

Nature Methods Points of Significance column: Statistics vs machine learning. (read)

Statistics asks us to choose a model that incorporates our knowledge of the system, and ML requires us to choose a predictive algorithm by relying on its empirical capabilities. Justification for an inference model typically rests on whether we feel it adequately captures the essence of the system. The choice of pattern-learning algorithms often depends on measures of past performance in similar scenarios.

Bzdok, D., Krzywinski, M. & Altman, N. (2018) Points of Significance: Statistics vs machine learning. Nature Methods 15:233–234.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: supervised methods. Nature Methods 15:5–6.

# Happy 2018 $\pi$ Day—Boonies, burbs and boutiques of $\pi$

Wed 14-03-2018

Celebrate $\pi$ Day (March 14th) and go to brand new places. Together with Jake Lever, this year we shrink the world and play with road maps.

Streets are seamlessly streets from across the world. Finally, a halva shop on the same block!

A great 10 km run loop between Istanbul, Copenhagen, San Francisco and Dublin. Stop off for halva, smørrebrød, espresso and a Guinness on the way. (details)

Intriguing and personal patterns of urban development for each city appear in the Boonies, Burbs and Boutiques series.

In the Boonies, Burbs and Boutiques of $\pi$ we draw progressively denser patches using the digit sequence 159 to inform density. (details)

No color—just lines. Lines from Marrakesh, Prague, Istanbul, Nice and other destinations for the mind and the heart.

Roads from cities rearranged according to the digits of $\pi$. (details)

The art is featured in the Pi City on the Scientific American SA Visual blog.

Check out art from previous years: 2013 $\pi$ Day and 2014 $\pi$ Day, 2015 $\pi$ Day, 2016 $\pi$ Day and 2017 $\pi$ Day.

# Machine learning: supervised methods (SVM & kNN)

Thu 18-01-2018
Supervised learning algorithms extract general principles from observed examples guided by a specific prediction objective.

We examine two very common supervised machine learning methods: linear support vector machines (SVM) and k-nearest neighbors (kNN).

SVM is often less computationally demanding than kNN and is easier to interpret, but it can identify only a limited set of patterns. On the other hand, kNN can find very complex patterns, but its output is more challenging to interpret.

Nature Methods Points of Significance column: Machine learning: supervised methods (SVM & kNN). (read)

We illustrate SVM using a data set in which points fall into two categories, which are separated in SVM by a straight line "margin". SVM can be tuned using a parameter that influences the width and location of the margin, permitting points to fall within the margin or on the wrong side of the margin. We then show how kNN relaxes explicit boundary definitions, such as the straight line in SVM, and how kNN too can be tuned to create more robust classification.

Bzdok, D., Krzywinski, M. & Altman, N. (2018) Points of Significance: Machine learning: a primer. Nature Methods 15:5–6.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

# Human Versus Machine

Tue 16-01-2018
Balancing subjective design with objective optimization.

In a Nature graphics blog article, I present my process behind designing the stark black-and-white Nature 10 cover.

Nature 10, 18 December 2017

# Machine learning: a primer

Thu 18-01-2018
Machine learning extracts patterns from data without explicit instructions.

In this primer, we focus on essential ML principles— a modeling strategy to let the data speak for themselves, to the extent possible.

The benefits of ML arise from its use of a large number of tuning parameters or weights, which control the algorithm’s complexity and are estimated from the data using numerical optimization. Often ML algorithms are motivated by heuristics such as models of interacting neurons or natural evolution—even if the underlying mechanism of the biological system being studied is substantially different. The utility of ML algorithms is typically assessed empirically by how well extracted patterns generalize to new observations.

Nature Methods Points of Significance column: Machine learning: a primer. (read)

We present a data scenario in which we fit to a model with 5 predictors using polynomials and show what to expect from ML when noise and sample size vary. We also demonstrate the consequences of excluding an important predictor or including a spurious one.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.