Sign in

Pierre-Paul Axisa

@ppaxisa.bsky.social
243 followers 137 following 338 posts

researcher at CBMR (Copenhagen). Genomics, genetics, compbio, biostats, pun lover. ppaxisa.github.io

PostsRepliesMedia
Pierre-Paul Axisa @ppaxisa.bsky.social · 3h
Trending towards average?
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 07/10/2026
The kids are alright
000
Pierre-Paul Axisa @ppaxisa.bsky.social · 06/10/2026
I remember an old discussion on LinkedIn about this, and problems related to confounding, and artefacts related to simulated data? With large sample sizes you will have several batches, I would model that. Also, from @wkhuber.bsky.social, weird properties of wilcoxon: bsky.app/profile/wkhu...
000
Pierre-Paul Axisa @ppaxisa.bsky.social · 06/10/2026
Add to the system prompt "on a pas gardé les cochons ensemble"
110
Pierre-Paul Axisa @ppaxisa.bsky.social · 05/10/2026
I’ve heard the same thing being done officiously in France (despite Inserm and CNRS adhering to DORA on paper). To add to the non sense, also weighting depending on author position and shared first/last; eg dividing the IF by number of co first authors, mechanically penalizing collaborative efforts😩
020
Pierre-Paul Axisa @ppaxisa.bsky.social · 05/10/2026
conflicted.r-lib.org #rstats
061
Pierre-Paul Axisa @ppaxisa.bsky.social · 04/10/2026
I don’t have any quantitative evidence but I think the mechanism might be stress induced by your partner calling: -where are you?! I’m at check out! -but I found this great Kirkland bundle, it’s on sale!
120
Pierre-Paul Axisa @ppaxisa.bsky.social · 04/10/2026
arrow+parquet has sped things up insanely for me recently
060
Reposted by Pierre-Paul Axisa
Pierre-Paul Axisa @ppaxisa.bsky.social · 03/10/2026
What do you think about AI generated text detection tools? I don’t know much about it practically. Reliable enough? Too costly anyway? Text can be partly AI generated but have sound reasoning anyway? Do you think it could be useful eg for random screens or when readers flag content?
101
Pierre-Paul Axisa @ppaxisa.bsky.social · 03/10/2026
What do you think about AI generated text detection tools? I don’t know much about it practically. Reliable enough? Too costly anyway? Text can be partly AI generated but have sound reasoning anyway? Do you think it could be useful eg for random screens or when readers flag content?
101
Pierre-Paul Axisa @ppaxisa.bsky.social · 01/10/2026
#academiclife
000
Reposted by Pierre-Paul Axisa
Erik van Nimwegen @erikvannimwegen.bsky.social · 01/10/2026
and completely opaque transformations to the data, and cannot be relied upon for almost anything. These extremely valuable datasets are often obtained at great costs and deserve that we take their analysis extremely seriously. We are not going to understand what these data tell us 13/n
111
Reposted by Pierre-Paul Axisa
Erik van Nimwegen @erikvannimwegen.bsky.social · 01/10/2026
The purpose of this thread is not to shit on the work of others but to defend our approach of being extremely careful and deliberate about methods. We are really concerned about typical standards in the single-cell omics field and the endless proliferation of methods that apply complex, ad hoc, 12/n
111
Reposted by Pierre-Paul Axisa
Erik van Nimwegen @erikvannimwegen.bsky.social · 01/10/2026
Hey, I heard about Bonsai, but now I heard there is another method called MILK that also visualizes scRNA-seq on trees. And it is much faster! Should I just use that then? A tree is a tree, right? TLDR: No. #methodsmatter Not even close. On realistic synthetic data MILK performs worse than UMAP.
1148
Pierre-Paul Axisa @ppaxisa.bsky.social · 29/09/2026
If you do a video bout it please call it "palette etiquette" 😀
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 24/09/2026
static.klipy.com
Dodgeball: It's a Bold Strategy, Cotton
ALT: Dodgeball: It's a Bold Strategy, Cotton
011
Reposted by Pierre-Paul Axisa
KLAB - finding a cure for GI dysmotility at #HarvardStrong @klab.bsky.social · 24/09/2026
4. Will we in academia wisen up and moderate our language while discussing our science, or will we match the hype-generation of AI cos by getting a bigger loudspeaker to generate more hype? What's at stake is nuance (which already was at risk in academics before advent of AI). (fin)
121
Pierre-Paul Axisa @ppaxisa.bsky.social · 22/09/2026
Makes sense, the AI is turning everything into bullet (points)
010
Reposted by Pierre-Paul Axisa
laserhedvig @laserhedvig.bsky.social · 21/09/2026
Interdisciplinary research is great and necessary, and also has consistently lower funding success www.nature.com/articles/nat... #academiclabourmarket #academia #research
26133
Pierre-Paul Axisa @ppaxisa.bsky.social · 20/09/2026
Could be very bad if it’s a requirement and viewed as a checkbox to publish, adding poor quality data to the record.
120
Pierre-Paul Axisa @ppaxisa.bsky.social · 20/09/2026
I wonder if the long term backlash to this will be that you will not be able to publish with just reanalysis. You’d have to generate new data to test your model to pass the publication filter. Could be positive loop if it adds good quality data to the record.
110
Pierre-Paul Axisa @ppaxisa.bsky.social · 19/09/2026
It drives me mad that changing research practices to improve reuse by humans, it gets limited traction. But when it’s for AI reuse, it’s all hands on deck.
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 18/09/2026
Great package name!
010
Reposted by Pierre-Paul Axisa
In Preparation @inprep.bsky.social · 18/09/2026
inpreparation.substack.com
Regev Daughter Bags Science Fair Win, Edges Baking Soda Volcano With Single-Cell Atlas Of Murine Tumor Microenvironment
SOUTH SAN FRANCISCO, CA—Genentech head of Research and Early Development Aviv Regev confirmed Tuesday that her 12-year-old daughter independently conceived, designed, and executed a 450,000-cell…
0136
Pierre-Paul Axisa @ppaxisa.bsky.social · 14/09/2026
Already several mentions of uses in bioinformatics like transcriptomics. little doubt that it’s useful but still a universal problem of how many. If you give several analyst the same data I think there could be a 10 fold variation on the # clusters "identified". This paper nicely asks "how many?"
doi.org
CHOIR improves significance-based detection of cell types and states from single-cell data - Nature Genetics
Cluster hierarchy optimization by iterative random forests (CHOIR) offers a robust and accurate method to identify cell clusters across a variety of single-cell resolution data with statistical suppor...
141
Pierre-Paul Axisa @ppaxisa.bsky.social · 14/09/2026
Damned it got me right in the intrinsically disordered region
020
Reposted by Pierre-Paul Axisa
Michael Love @mikelove.bsky.social · 12/09/2026
I wrote a post about how slice_sample() and other slice_* functions work in #tidyomics #Bioconductor tidyomics.github.io/tidyomicsBlo...
tidyomics.github.io
Slicing in tidyomics – tidyomicsBlog
How to use dplyr-style slice operations on omics data objects in the tidyomics project.
02213
Pierre-Paul Axisa @ppaxisa.bsky.social · 08/09/2026
I was going to say, you can snooze an alarm, you can’t snooze a toddler.
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 08/09/2026
Does toddler count as an alarm?
121
Reposted by Pierre-Paul Axisa
Pierre-Paul Axisa @ppaxisa.bsky.social · 08/09/2026
Since scRNAseq happened I have heard many times "we don’t need large sample sizes, we have SiNgLe cells"
static.klipy.com
That's Not How It Works! That's Not How Any Of This Works!
ALT: That's Not How It Works! That's Not How Any Of This Works!
231
Pierre-Paul Axisa @ppaxisa.bsky.social · 08/09/2026
💯 to everything. I love your "read the documentation" recurring posts. I would add an extra nugget: more than once I go read the documentation for something specific, and as a side effect discover a cool feature that I will use the next time around. (And I browse all the repos of the dev for 1 hour)
060
Pierre-Paul Axisa @ppaxisa.bsky.social · 08/09/2026
Since scRNAseq happened I have heard many times "we don’t need large sample sizes, we have SiNgLe cells"
static.klipy.com
That's Not How It Works! That's Not How Any Of This Works!
ALT: That's Not How It Works! That's Not How Any Of This Works!
231
Reposted by Pierre-Paul Axisa
Rachael Bay @rachaelabay.bsky.social · 05/09/2026
I have recently noticed a string of vandalism on our floor. These outrageous acts seem to be correlated with visits from a certain corn geneticist @jrossibarra.bsky.social
1223
Reposted by Pierre-Paul Axisa
Ewan Birney @ewanbirney.bsky.social · 04/09/2026
Two weeks ago the @economist.com had a leader "In praise of designer-ish babies". I wrote a letter to the Editor about this, explaining why the UK system is so confident that this should remain unlawful. The Economist has kindly published a shortened form+ my full letter below lnkd.in/p/ef-n78N8
lnkd.in
Full letter to Economist, response to "In praise of designer-ish babies" | Ewan Birney
Two weeks ago the The Economist published a leader titled "In praise of designer-ish babies". I wrote a letter to the editor in response to this. The Economist was kind enough to publish a shortened f...
2125
Pierre-Paul Axisa @ppaxisa.bsky.social · 04/09/2026
I have slowly moved to using AnnotationHub, which I think has become the recommended approach in the bioconductor ecosystem. At least what I remember from going to CSAMA workshop few years ago.
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 29/08/2026
I’m not at all used to working in containers but I found the extensions in VScode/Positron make it a breeze to get started. I work mostly in R so I started from the Rocker project, then customized to might needs: rocker-project.org. Then let the LLM run wild in the container.
rocker-project.org
The Rocker Project – Rocker Project
Docker Containers for the R Environment
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 27/08/2026
I am getting dizzy with all the flashy LLM generated diagrams and charts in scientific communication. I want to understand what your research is about, a clear title and an odd box and arrows will do perfectly well. I remember nostalgically my postdoc days where lab meetings were chalk only.
010
Reposted by Pierre-Paul Axisa
Erik van Nimwegen @erikvannimwegen.bsky.social · 27/08/2026
Want to explore and understand the structure of your single-cell omics or other high-dimensional data? Try Bonsai! www.nature.com/articles/s41... Check out our tutorial videos with 1. Highlights of the Bonsai-scout exploratory analysis tool: www.youtube.com/watch?v=TE7T...
youtube.com
Bonsai-scout: Highlights
YouTube video by ISMARA: inferring gene regulatory interactions
1135
Pierre-Paul Axisa @ppaxisa.bsky.social · 27/08/2026
If you’re looking for something with publications. I also looked through @wkhuber.bsky.social lab website recently: github.com/Huber-group-...
github.com
GitHub - Huber-group-EMBL/Huber-group-EMBL.github.io: Huber group public webpage
Huber group public webpage. Contribute to Huber-group-EMBL/Huber-group-EMBL.github.io development by creating an account on GitHub.
120
Pierre-Paul Axisa @ppaxisa.bsky.social · 27/08/2026
@nrennie.bsky.social one is great nrennie.rbind.io
nrennie.rbind.io
Nicola Rennie
120
Pierre-Paul Axisa @ppaxisa.bsky.social · 26/08/2026
What seed did you use 🤓?
291
Pierre-Paul Axisa @ppaxisa.bsky.social · 26/08/2026
Who is left? CST are the only ones being reliable?
010
Reposted by Pierre-Paul Axisa
Biozentrum, University of Basel @biozentrum.unibas.ch · 21/08/2026
How do you visualize highly complex data without losing important information? Prof. Erik van Nimwegen’s team developed a new tool, Bonsai, that provides a faithful picture of the structure in the data. @erikvannimwegen.bsky.social @unibas.ch @dhdegroot.bsky.social @imbavienna.bsky.social
biozentrum.unibas.ch
Seeing the Unseen: A new tool for exploring complex biological data
Modern biology research can now generate valuable data on an unprecedented scale. But there is one major obstacle: making sense of these complex high-dimensional datasets. Prof. Erik van Nimwegen’s te...
0258
Reposted by Pierre-Paul Axisa
jp flores (he/him) @jpflores.rbind.io · 19/08/2026
1/7 Prestige sucks. Not because recognition is bad. But because science has spent decades confusing journal prestige with scientific quality. And that shapes what gets published, what gets rewarded, and ultimately what science gets done.
1187
Pierre-Paul Axisa @ppaxisa.bsky.social · 20/08/2026
All in the browser?! This is freaking amazing! I don’t run flow experiments anymore but if there’s one thing I don’t miss it’s flowjo
110
Reposted by Pierre-Paul Axisa
vincentholst.bsky.social @vincentholst.bsky.social · 14/08/2026
As promised, we now share some details why it took 32 months to get this Matters Arising published, and, since many asked, share some comments on the reply to our critique. All details can be found in this small FAQ vincentholst.github.io/the_curious_.... Below is a summary 🧵
vincentholst.github.io
The Curious Case of the Declining Disruption’s Disappearance
48348
Pierre-Paul Axisa @ppaxisa.bsky.social · 12/08/2026
The argument I have heard most consistently is that you gain nothing from it and you risk scooping (Not endorsing). Mostly from experimental persons, eg mouse model systems that hard to get to work but easy to do once you know the parameters
010
Pierre-Paul Axisa @ppaxisa.bsky.social · 07/08/2026
Support from the department chair is important
100
Reposted by Pierre-Paul Axisa
Dorothy Bishop @deevybee.bsky.social · 21/07/2026
I'm attending online event about a #NIH competition to develop a data-sharing index (S-Index), but am surprised that there aren't more people here (it's hybrid). www.nei.nih.gov/sindex Nice idea for NIH to have a competition for developing a S-index; we've just heard about the winners. 1/n
nei.nih.gov
NIH S-Index Innovation Event | National Eye Institute
Registration: http - July 21, 2026
263
Pierre-Paul Axisa @ppaxisa.bsky.social · 20/07/2026
static.klipy.com
Elrond's Famous Line: I Was There Gandalf
ALT: Elrond's Famous Line: I Was There Gandalf
110