Månadsarkiv: maj 2020

Better posters are nice, but we need better poster session experiences

Publicerat i english, the practice of science av

Fear and loathing in the conference centre lobby

Let me start from a negative place, because my attitude to poster sessions is negative. Poster sessions are neither good ways to communicate science, nor to network at conferences. Moreover, they are unpleasant.

The experience of going to a poster session, as an attendant or a presenter goes something like this: You have to stand in a crowded room that is too loud and try to either read technical language or hold a conversation in about a difficult topic. Even without anxiety, mobility, or hearing difficulties, a poster session is unlikely to be enjoyable or efficient.

Poster sessions are bad because of necessities of conference organisation. We want to invite many people, but we can’t fit in many talks; we get crowded poster sessions.

They are made worse by efforts to make them better, such as mandating presenters to stand by their posters, in some cases on pain of some sanction by the organisers, or to have the poster presenters act as dispensers of alcohol. If you need to threaten or drug people to participate in an activity, that might be a sign.

They are made not worse but a bit silly, by assertions that poster sessions are of utmost importance for conferencing. Merely stating that the poster session is vibrant and inspiring, or that you want to emphasise the poster as an important form of communication, sadly, does not make it so, if the poster sessions are still business as usual.

Mike Morrison’s ”Better Scientific Poster” design

As you can see above, my diagnosis of the poster session problem is part that you’re forced to read walls of text or listen to mini-lectures, and part that it happens in an overcrowded space. The walls of text and mini-lecture might be improved by poster design.

Enter the Better Scientific Poster. I suggest clicking on that link and looking at the poster templates directly. I waited too long to look at the actual template files, because I expected a bunch of confusing designer stuff. It’s not. They contain their own documentation and examples.

There is also a video on YouTube expanding on the thinking behind the design, but I think this conversation on the Everyting Hertz podcast is the best introduction, if you need an introduction beyond the template. The YouTube video doesn’t go into enough detail, and is also a bit outdated. The poster template has gone through improvements since.

If you want to hear the criticisms of the design, here’s a blog post summarising some of it. In short, it is unscientific and intellectually arrogant to put a take home message in too large a font, and it would be boring if all posters used the same template. Okay.

The caveats

I am not a designer, which should be abundantly clear to everyone. I don’t really know what good graphic design principles for a poster are.

There is also no way to satisfy everyone. Some people will think you’ve put too little on the poster unless it ”tells the full story” and a has self-contained description of the methods with all caveats. Some people, like me, will think you’ve put way too much on it long before that.

What I like, however, is that Morrison’s design is based on an analysis of the poster session experience that aligns with mine, and that it is based on a goal for the poster that makes sense. The features of the design flow from that goal. If you listen to the video or the Hertz episode: Morrison has thought about the purpose of the poster.

He’s not just expressing some wisdom his PhD supervisor told him in stern voice, or what his gut feeling tells him, which I suspect is the two sources that scientists’ advice on communication is usually based on. We all think that poster sessions are bad, because we’ve been to poster sessions. We usually don’t have thought-through ideas about what to do better.

Back to a place of negativity

For those reasons, I think the better poster is likely to be an improvement. I was surprised that I didn’t see it sweep through poster sessions at the conferences I went to last summer, but there were a few. I was going to try it for TAGGC 2020 (here is my poster aboutthe genetics of recombination rate in the pig), but that moved online, which made poster presentations a little different.

However, changing up poster layout can only get you so far. Unless someone has a stroke of genius to improve the poster viewing experience or change the economics of poster attendance, there no bright future for the poster session. Individually, the rational course of action isn’t to fiddle with the design and spend time to squeeze marginal improvements out of our posters. It is to spend as little time as possible on posters, ignoring our colleagues’ helpful advice on how to make them prettier and more scientific, and lowering our expectations.

Robertson on genetic correlation and loss of variation

Publicerat i citerat, english, genetik av

It’s not too uncommon to see animal breeding papers citing a paper by Alan Robertson (1959) to support a genetic correlation of 0.8 as a cut-off point for what is a meaningful difference. What is that based on?

The paper is called ”The sampling variance of the genetic correlation coefficient” and, as the name suggests, it is about methods for estimating genetic correlations. It contains a section about the genetic correlation between environments as a way to measure gene-by-environment interaction. There, Robertson discusses experimental designs for detecting gene-by-environment interaction–that is, estimating whether a genetic correlation between different environments is less than one. He finds that you need much larger samples than for estimating heritabilities. It is in this context that the 0.8 number comes up. Here is the whole paragraph:

No interaction means a genetic correlation of unity. How much must the correlation fall before it has biological or agricultural importance? I would suggest that this figure is around 0.8 and that no experiment on genotype-environment interaction would have been worth doing unless it could have detected, as a significant deviation from unity, a genetic correlation of 0.6. In the first instance, I propose to argue from the standpoint of a standard error of 0.2 as an absolute minimum.

That is, in the context of trying to make study design recommendations for detecting genotype-by-environment interactions, Robertson suggests that a genetic correlation of 0.8 might be a meaningful difference from 1. The paper does not deal with designing breeding programs for multiple environments or the definition of traits, and it has no data on any of that. It seems to be a little bit like Fisher’s p < 0.05: Suggest a rule of thumb, and risk it having a life of its own in the future.

In the process of looking up this quote, I also found this little gem, from ”The effect of selection on the estimation of genetic parameters” (Robertson 1977). It talks about the problems that arise with estimating genetic parameters in populations under selection, when many quantitative genetic results, in one way or another, depend on random mating. Here is how it ends:

This perhaps points the moral of this paper. The individuals of one generation are the parents of the next — if they are accurately evaluated and selected in the first generation, the variation between families will be reduced in the next. You cannot have your cake and eat it.

Literature

Robertson, A. ”The sampling variance of the genetic correlation coefficient.” Biometrics 15.3 (1959): 469-485.

Robertson, A. ”The effect of selection on the estimation of genetic parameters.” Zeitschrift für Tierzüchtung und Züchtungsbiologie 94.1‐4 (1977): 131-135.

Using R: setting a colour scheme in ggplot2

Publicerat i computer stuff, data analysis, english av

Note to self: How to quickly set a colour scheme in ggplot2.

Imagine we have a series of plots that all need a uniform colour scale. The same category needs to have the same colour in all graphics, made possibly with different packages and by different people. Instead of hard-coding the colours and the order of categories, we can put them in a file, like so:

library(readr)
colours <- read_csv("scale_colours.csv")

# A tibble: 5 x 2
  name   colour 
      
1 blue   #d4b9da
2 red    #c994c7
3 purple #df65b0
4 green  #dd1c77
5 orange #980043

Now a plot with default colours, using some made-up data:

x <- 1:100

beta <- rnorm(5, 1, 0.5)

stroop <- data.frame(x,
                     sapply(beta, function(b) x * b + rnorm(100, 1, 10)))
colnames(stroop)[2:6] <- c("orange", "blue", "red", "purple", "green") 

data_long <- pivot_longer(stroop, -x)

plot_y <- qplot(x = x,
                y = value,
                colour = name,
                data = data_long) +
  theme_minimal() +
  theme(panel.grid = element_blank())

Now we can add the custom scale like this:

plot_y_colours <- plot_y + 
  scale_colour_manual(limits = colours$name,
                      values = colours$colour)