Q: What's wrong with these plots? A: Pooling cell-level measurements across multiple runs conceals experimental reproducibility. Read more in JCB (with @fritzlaylin.bsky.social @katrinavelle.bsky.social & Dyche Mullins): rupress.org/jcb/article/... - ThreadSky

samjlord.bsky.social • 102 days ago

Q: What's wrong with these plots?

A: Pooling cell-level measurements across multiple runs conceals experimental reproducibility.

Read more in JCB (with @fritzlaylin.bsky.social @katrinavelle.bsky.social & Dyche Mullins): https://rupress.org/jcb/article/219/6/e202001064/151717/SuperPlots-Communicating-reproducibility-and

1 / 2

Comments

poppi62.bsky.social•102 days ago

We have made superplots our standard or a while now

michellejansman.bsky.social•102 days ago

I am going to give this a try too (also in not cell-related experiments), it gives such a clear view of the complete data set.

Q: What would you say is min. amount of data points within the n=1 for trying SuperPlots?

samjlord.bsky.social•101 days ago

I would say as long as you have at least 3 subsample measurements, it should look reasonable. Once you have thousands of datapoints, it makes sense to use violin plots instead of scatter plots for the subsample points. :)

ffchcarvalho.bsky.social•102 days ago

SuperPlots are awesome at conveying all the experimental replication data in a visually simple and very clear manner 👍

annbajur.bsky.social•102 days ago

Seeing this makes me wanna 😢 I see this way too often!

katecavanaugh.bsky.social•102 days ago

Superplots FTW!

westwingslab.bsky.social•102 days ago

i show this to everyone in my lab. We've started using superplots to show individual Drosophila overlayed with the mean of the individual experimental cross (true biological replicate)

peiferlabunc.bsky.social•102 days ago

Love these and its great they made it simple to generate them--we've switched over

patchyclamp.bsky.social•101 days ago

This is really great, thanks for this!

astormic.bsky.social•102 days ago

A very good thread! Thanks Sam!

samjlord.bsky.social•102 days ago

Note that all the different experimental scenarios on the right—including random flask-to-flask variability in D—can yield the same plots and tiny error bars in A when we treat each cell as an "n":

drewfriedmann.bsky.social•102 days ago

I love these plots and make anyone working with me read your paper (along with this one: https://www.nature.com/articles/nn.2886). But mostly I love that "STILL NOPE" made it into the final figure.

samjlord.bsky.social•102 days ago

Our editor at @jcellbiol.bsky.social was awesome.

samjlord.bsky.social•102 days ago

So we proposed #SuperPlots, which superimpose sample-to-sample summary data on top of the cell-level distribution. It's a simple way to convey both variability of the underlying data and the repeatability of the experiment. Here, sample means are overlaid on a violin plot of all the cell data.

samjlord.bsky.social•102 days ago

You an even go as far as to color-code each cell-level datapoint by the experimental run it came from, so the reader can see if there if the cell variability is clustered or dispersed.

samjlord.bsky.social•102 days ago

We have tutorials on how to generate SuperPlots in Prism, R, Python, and Excel in the paper, but you can also easily make SuperPlots using @joachimgoedhart.bsky.social's SuperPlotsOfData: https://huygens.science.uva.nl/SuperPlotsOfData/

anntivey.bsky.social•102 days ago

📌

bertolingiulia.bsky.social•102 days ago

Would you have a link towards a tutorial on how to generate Superplots in Prism? Asking for… me 😅

samjlord.bsky.social•102 days ago

It’s in the supplement for the original paper (Figure S2). Also, a video tutorial here: https://youtu.be/VKsAHUGvfTE?si=u56pd2dLJAkidhqz

jeffmorgan1967.bsky.social•102 days ago

📌

ffchcarvalho.bsky.social•102 days ago

📌

greenrat.bsky.social•102 days ago

📌

marcodv.bsky.social•102 days ago

I love SuperPlots! If I can shamelessly share an article I wrote this year about some do's and don'ts of data viz, I also included a reproduction of the same figures of the paper in Python here: https://pub.ista.ac.at/~mdallave/read/viz-dos-donts/main/#superplots

samjlord.bsky.social•102 days ago

Written tutorials are in the supplementary figures. And a video tutorial for Prism here: https://youtu.be/VKsAHUGvfTE?si=u56pd2dLJAkidhqz

samjlord.bsky.social•102 days ago

There are also cool Violin SuperPlots by Martin Kenny and Ingmar Schoen: https://www.molbiolcell.org/doi/10.1091/mbc.E21-03-0130

zs-biophys.bsky.social•80 days ago

Thanks @andreamarat.bsky.social! Have been looking for this paper in the past several days!

andreamarat.bsky.social•79 days ago

I love seeing things SuperPlotted! also really like this paper on calculating how many cells you actually need to average https://rupress.org/jcb/article/223/8/e202401074/276824/Determining-how-many-cells-to-average-for?searchresult=1

Posting Rules

Be respectful to others
No spam or self-promotion
Stay on topic
Follow Bluesky's terms of service

Comments

Posting Rules

Reply