Pablo Riesgo-Ferreiro

Protein language models excel at generating functional yet remarkably diverse artificial sequences.

They however fail to naturally sample rare datapoints, like very high activities.

In our new preprint, we show that RL can solve this without the need for additional data:

arxiv.org/abs/2412.12979

Spatial transcriptomic clocks reveal cell proximity effects in brain ageing Nature - A spatially resolved single-cell transcriptomics map of the mouse brain at different ages reveals signatures of ageing, rejuvenation and disease, including ageing effects associated with T...

Thrilled to see this paper out!! 🧪

Spatial transcriptomics of brain aging and 'spatial aging clocks' identify cells that have pro-aging or pro-rejuvenating effects on their neighbors!

Huge CONGRATS to Eric Sun and all authors! Fantastic collaboration with @jameszou.bsky.social!

rdcu.be/d33uQ 🧵

Cell is happy to connect (and re-connect) with scientists across all areas of biology on Bluesky! We aim to start posting in 2025. Until then, we'll be keeping an eye out for your exciting and interesting findings... 🔬👀

One of the biggest questions we get is when will our journals be here. We thank you for your patience.

You can now follow our journals with this starter pack!

Infamous paper that popularized unproven COVID-19 treatment finally retracted Study on hydroxychloroquine by Didier Raoult and colleagues gets pulled on ethical and scientific grounds

It took four days from submission to publication, and nearly five years from publication to retraction. After campaigning by many, many scientists, and an investigation by Elsevier, an infamous paper on hydroxychloroquine as a COVID-19 treatment has been retracted. 🧪
www.science.org/content/arti...

Genome in a Bottle Consortium goals:The Genome in a Bottle Consortium is a public-private-academic consortium hosted by NIST to develop the technical infrastructure (reference standards, reference methods, and reference...

Maybe keep an eye on the genome in a bottle initiative www.nist.gov/programs-pro...

The RNA part is less mature than others but still you'll find there some public human samples which you can analyse and contrast results.

The 2025 EMBL Annual Poster, listing all of the courses and conferences taking place in 2025 at EMBL

Hi Bluesky community! 👋

Are you looking forward to 2025 as much as we are? 👀

We just updated our 2025 Annual Poster to include even more events for you to join, so go and have a look!

➡️ s.embl.org/poster-bl

#EMBLEvents #molecularbiology #lifesciencetraining

Calling Bullshit: Data Reasoning in a Digital World The world is awash in bullshit. Politicians are unconstrained by facts. Science is conducted by press release. Higher education rewards bullshit over analytic thought. Startup culture elevates bullshi...

As I mentioned three weeks ago, Jevin West and I are developing a sort of followup course to Calling Bullshit.

We've got the basic scrollytelling tech worked out and now need to choose a design.

We're not designers and would like to simply buy an HTML5 template if only for the colors and fonts.

«Científico», una palabra inventada por españoles que en primer lugar se usó para hablar de una mujer La palabra apareció en un libro de Juan de Mena en el siglo XVI y fue popularizada por William Whewell en el XIX. La primera persona descrita por ella fue Mary Somerville

Interesting! From what I find it was in a book by Juan de Mena, las Trezientas in 1566 and later on by Manuel Martin in 1788.

Or that says here:
www.abc.es/ciencia/abci...

What if Eroom's law, the decreasing productivity of drug discovery, is a result of aggressive accounting whereby pharma companies consider as much stuff as possible as R&D to avoid taxes? I bet eroom's law holds in movies too - where losses are built to equal move revenue

EmbedGEM: a framework to evaluate the utility of embeddings for genetic discovery AbstractSummary. Machine learning-derived embeddings are a compressed representation of high content data modalities. Embeddings can capture detailed infor

In our work (EmbedGEM: doi.org/10.1093/bioa...) we developed the first formal evaluation framework for multivariate phenotypes that reports metrics for both heritability as well as disease relevance. The way we evaluate heritability is using several simple proxies: number of associations, (6/)

This is possibly the last thing I was missing from R in python. Thanks for that!

🎉Thrilled that #GenePT is published!

We show how to create #ChatGPT embeddings of genes and #singlecell --> improve gene function prediction + cell annotation/analyses🧬
📰 www.nature.com/articles/s41...
🌐 github.com/yiqunchen/Ge...
Great job Yiqun Chen! #LLM for bio.

Imperfect wound healing sets the stage for chronic diseases Although the age of the genome gave us much insight about how our organs fail with disease, it also suggested that diseases do not arise from mutations alone; rather, they develop as we age. In this R...

Out in time for the #ASCB meeting an @science.org Review from Jody Rosenblatt and friends
www.science.org/doi/10.1126/...

The altmetric works for bluesky!

Frontiers | Immune-based subgroups uncover diverse tumor immunogenicity and implications for prognosis and precision therapy in acute myeloid leukemia BackgroundAlthough a considerable proportion of acute myeloid leukemia (AML) patients achieve remission through chemotherapy, relapse remains a recurring and...

www.frontiersin.org/journals/imm...

An approach to stratify AML patients in two immune groups based on expression profiles.

#paper #immunology

The one thing that I still do not find satisfactory in VScode which pycharm excels at is code refactoring. There are so many patterns in pycharm... I hope they implement something close soon in VScode

Hi @benblaiszik.bsky.social would you, please, add me here? AI for immunology and Bioinformatics in general

Thanks @kevinkaichuang.bsky.social would you add me there too?

Why don't you just use an ssh remote? The code will be stored remotely but editing is "local". 2, 3 and 4 are met.
This is precisely something that pycharm does but only with the pro version not with the free version.

This is so arbitrary it's laughable in retrospective

Sketch of a "pterodactyl" labelled as Archeopter by Nicholas Miklouho-Maclay

Doodle below writing of a plesiosaur and Pterodactyl

Ammonite anatomical drawings

Imagine your notebook surviving for centuries to be studied by historians. This happened to transcriptions by Nicholai Miklouho-Maclay, Russian of Ukrainian Cossack descent, of a palaeontology class delivered by Ernst Haeckel (1866) This unlikely record, captures the rise of race science in Germany🧵

What Are Shapley Interactions, and Why Should You Care? A guest post by Julia, Max, Fabian and Hubert.

What are shapley interactions and why should you care about them?

This is a guest post by Julia, Max, Fabian and Hubert on my newsletter Mindful Modeler.

I also learned a lot from this post and definitely recommend checking out the shapiq package.

mindfulmodeler.substack.com/p/what-are-s...

Timescales in Cell Biology #Cell

PDF download:
www.cell.com/cell/pdf/S00...

Average token length

Domain boundary hit percentage

Number of distinct neighbors each token encounters

A nice analysis of different tokenization strategies (BPE, wordpiece, sentencepiece) on protein sequences.

arxiv.org/abs/2411.17669

In case some mentors may want to read this. #Nature #PhD #scisky #student #mentor

Source:
www.researchgate.net/publication/...

Design of high specificity binders for peptide-MHC-I complexes https://www.biorxiv.org/content/10.1101/2024.11.28.625793v1 Class I MHC molecules present peptides derived from intracellular antigens on the cell surface for i

Design of high specificity binders for peptide-MHC-I complexes https://www.biorxiv.org/content/10.1101/2024.11.28.625793v1

Our Herpesfolds paper is out in Nature Communications! rdcu.be/d1uM7 Congrats to Timothy Soh,
@sofiaaognibene
, Saskia Sanders, Robin Schaeper and Benedikt Kaufer!
Have a look at our database at herpesfolds.org by Malte Kansy

A panoramic view of cell population dynamics in mammalian aging To elucidate aging-associated cellular population dynamics, we present PanSci, a single-cell transcriptome atlas profiling over 20 million cells from 623 mouse tissues across different life stages, se...

Immune cells change in proportions with age - naive go down, effector go up. www.science.org/doi/10.1126/...

The danger theory of immunity strikes back... rdcu.be/d1uQN