This is actually a pretty neat idea. Spatially correlated genes.
www.biorxiv.org/content/10.1...
@pcnmartin.bsky.social
Aiming to make the Solar Punk Aesthetic a reality. Science, World Building and Story telling. Researcher/Bioinformatician/Data Scientist. "Always strive to contribute to others." Contact: https://patrickcnmartin.github.io/
This is actually a pretty neat idea. Spatially correlated genes.
www.biorxiv.org/content/10.1...
Well. I'm not going to say I am surprised by these results. Come on people. We can do better than this.
www.nature.com/articles/s41...
#AI #Honesty
Often working at the intersection of AI and Biology, I have developed a love-hate relationship with AI.
In this essay, I share some thoughts on the modern AI ecosystem. Itβs not all doom and gloom but there is room for improvement.
pcnmartin.substack.com/p/a-love-hat...
Vesalius 2.0 is in βͺβͺ@natcomms.nature.comβ¬ !
We show how using spatial context improves cell-to-cell mapping across heterogenous spatial samples.
Thanks to the reviewers, editors, and collaborators for making this possible.
GitHub: github.com/WonLab-CS/Ve...
Paper: www.nature.com/articles/s41...
It has been quite fun and interesting trying to run scRNA analysis using only distributed/parallelized packages. Finding approximations to non-parallelizable algorithms is quite the challenge.
It also makes you appreciate the amazing work behind Scanpy from the @scverse.bsky.social and Seurat.
Thank you @qchengx1e3.bsky.social for handling this manuscript! Great to see it out.
22.08.2025 14:29 β π 3 π 0 π¬ 0 π 0Vesalius 2.0 is in βͺβͺ@natcomms.nature.comβ¬ !
We show how using spatial context improves cell-to-cell mapping across heterogenous spatial samples.
Thanks to the reviewers, editors, and collaborators for making this possible.
GitHub: github.com/WonLab-CS/Ve...
Paper: www.nature.com/articles/s41...
Does this plot look ugly? Yes it does.
Am I happy with because of what it represents? Yes I am.
Simulated Gene Pulse
Want to read an article but the article is behind a paywall? I have a trick to get around that.
It is a bit convoluted but hear me out:
1. Open your email.
2. Get the email of the corresponding authors
3. Send them a polite email that you wrote yourself asking for a pdf of the paper.
4. Profit
I am not sure any of those questions would require a token greedy language bot to answer.
11.08.2025 19:22 β π 1 π 0 π¬ 0 π 0My first "officially" published #scifi story in @naturefutures.bsky.social, a venue which has published stories by the likes of Arthur C Clark and Kim Stanley Robinson!
The story is entitled: A rude awakening. Just breath.
www.nature.com/articles/d41...
My first "officially" published #scifi story in @naturefutures.bsky.social, a venue which has published stories by the likes of Arthur C Clark and Kim Stanley Robinson!
The story is entitled: A rude awakening. Just breath.
www.nature.com/articles/d41...
The voyage was over. After 300 years in stasis, the crew had reached their new home. It was time to wake up. To start living again. Only β¦ something wasnβt right. They felt different. What had happened while they slept??
#scifi by @pcnmartin.bsky.social www.nature.com/articles/d41...
Being a researcher requires a certain level of masochism. You have to like to humiliate yourself by feeling stupid everyday.
You read some paper and you think: "Wow this actually makes sense to some people. The equations look cool I guess."
You know how ChatGPT became a bit of a sycophant, validating everyone thoughts, and finding ways to justify them?
How do we know that Agentic AI systems performing a task will not do the same thing? Will they start p-hacking? Cherry picking data?
#LLM #AgenticAI #AI
I have a suggestion but it comes with a serious caveat: I am not sure it was ever translated to english. If it ever does get translated, I would highly recommend:
The Horde of The Counterwind by Alain Damasio
I love attending conferences but I often struggle to stay attentive during scientific presentations. The main reason: these talks are dryer than Arrakis ever was.
Here, I share some thoughts on scientific story telling through the lens of fiction.
open.substack.com/pub/pcnmarti...
The Arc Institute has just released their pre-print STATE.
"Predicting cellular responses to perturbation
across diverse contexts with State"
arcinstitute.org/manuscripts/...
I love attending conferences but I often struggle to stay attentive during scientific presentations. The main reason: these talks are dryer than Arrakis ever was.
Here, I share some thoughts on scientific story telling through the lens of fiction.
open.substack.com/pub/pcnmarti...
I have just read the transcript of an interview with Robert Taylor (head of ARPA).
He had some interesting things to say about AI and the issue with AI hype. This was in 1989 and still holds wisdom today.
conservancy.umn.edu/items/16d6b5...
It was just a matter of time before we saw the rise of AI charlatans. Not sure it is worse than AI generated rage bait though...
03.06.2025 22:18 β π 0 π 0 π¬ 0 π 0Through some conversations with ChatGPT, it said this about Belgium:
"Belgium is like the Monty Python sketch of nations: serious face, absurd punchline."
As a Belgian, I can't say I disagree.
This is an extremely powerful method! In essence, they can re-create the raw data from embedding space across models.
arxiv.org/abs/2505.12540
After a month of developing with #Julia, it is becoming my favorite language! It's crazy how having strict type setting can catch most bugs. Yes, I know python has that as well. And Rust. And C/C++. But R doesn't...
New project coming out soon? Any guesses as to what this is?
"ML is the process of learning from data to perform
complex tasks for which there is no known deterministic and algorithmic solution, or building such
a solution is not practical."
Not all problems need ML. Not all questions can be solved without. Choose Wisely.
arxiv.org/abs/2312.03120
I am so tempted to say St Anger but that would be a lie. Enter Sandman.
29.04.2025 22:09 β π 0 π 0 π¬ 0 π 0You know what? Package development in Julia is pretty neat.
#Julia #Dev
The Institute
Chapter 4: Rock Vocals
The single link between geology and singing
pcnmartin.substack.com/p/rock-vocals
In that last few weeks, I have been seeing a few posts talking about the caveats of "vibe coding". Unsurprisingly, "vibe coding" only works if you are already an expect programmer who can spot terrible code at a glance.
So I ask: What makes us think that PhD level agents are a good idea ?
Well the Evo 2 model from the @arcinstitute.org has been dropped. Looking forward to reading this one.
arcinstitute.org/manuscripts/...