🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

A spatial transcriptomics method that… 🧪is easy to use 🔬has high resolution 🪙is cost-efficient 🌍scales to 3D Just in time for Xmas: Open-ST, a end-to-end, open-source method that checks all boxes, from the Nikolaus Rajewsky lab 📄Preprint: tiny.cc/openst 🧵⬇️

I want to apologize for the tone of my comments in this network in the past few weeks. They weren't directed to anyone in particular, and I'm sorry if I've offended anyone. Science is about advancing knowledge, and that can only be achieved by a united research community.

Our trash panda made it to cover for the new issue of Nature Neuroscience, highlighting our recent paper on the circuits encoding circadian feeding time. Artwork by Muhammed İkbal Alp

Every so often, the question of whether to normalize RNA counts by cell volume comes up in lab. Default answer: "yes, normalize by volume". Why? It's not just that RNA correlates with volume. It's that the relationship is *causal*. If you increase volume, RNA goes up. Thread 1/

We compared six commercial + academic multiplexed in-situ gene expression profiling technologies. Check out our preprint, led by Austin Hartman. Takeaway when evaluating spatial tech: consider both sensitivity and specificity! biorxiv.org/cgi/content/sh…

It's there! scCensus: Off-target scRNA-seq reads reveal meaningful biology biorxiv.org/content/10.110… An huge effort led by Dongze He! He seeks to answer "what's with all these #scRNA-seq reads that aren't mapping as expected?". I'll ask him to condense key points for a tweetorial.

When you can tell UMAPs were generated with PCA preprocessing just by looking at them... Other than that, this seems like an amazing tool!

ML/SystemBio/Single-cell folks: Which single characteristic weighs the most when your group decides to test/adopt a new bioinformatics or ML tool? Trying to understand what makes a tool widely used. Asking for a friend if anyone asks

Our paper on differentiable marker selection is finally out! nature.com/articles/s4154… - it was a great team effort over many years with Soledad Villar

A major update for DataMapPlot adds interactive plots. See lmcinnes.github.io/datamapplot_ex… for an example. Let's dig in to what you can do with DatMapPlot 0.2 ... 🧵

It's sad to see respectable scientists being so opinionated about an algorithm they can't begin to understand. It seems that "political correctness", appearances and pedigree have taken over actual, objective, factual knowledge. Now, if you say a word against using PCA... 🙃

🎉ICLR 2026: Simplicial Representation Learning w/ Neural k-Forms #Geometry #MachineLearning #ICLR2024 📜openreview.net/forum?id=Djw0X… TL;DR: Learnable differential forms = ꜰᴇᴀᴛᴜʀᴇ ᴍᴀᴘꜱ for combinatorial complexes (like graphs), enabling classification + more! 🧵1/n

In their Specious Art paper, Chari & Lior Pachter claim that tSNE/UMAP are as arbitrary as a random elephant shape. But are they? We show in our comment that this is false and throws the tSNE/UMAP baby out with the bathwater! Details in 🧵& paper: biorxiv.org/content/10.110… 1/8

To the colleagues and friends who find our paper overwhelming, and instead prefer to write and read more manageable papers with more focused messages: I respect your perspective and your approach to science greatly, and find it invaluable, of course. Both broad and focused

Friends, I am beyond happy! I'm starting a new position as Full Professor of #MachineLearning at the University of Fribourg Uni Fribourg 🇨🇭! With #SwissAI and many other initiatives, I am taking my research at the intersection of #geometry, #topology, and #MachineLearning to a new

DPAG's Domingos Group has a new paper published nature - Understanding the Biological Basis of Obesity: A Sympathetic Path to Treatment by burning fat dpag.ox.ac.uk/news/new-paper…

Check out Yifan's thread on the computational details of Spateo on 3D alignment, reconstruction and how this same framework can be used to reconstruct cell migration vector field to connect macroscopic morphogenesis with microscopic molecular dynamics. Yifan Lu and I will be