🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣 ….. oh my sides ….. 🤣🤣🤣🤣🤣🤣🤣🤣

In David Attenborough voice: Here we see an authentic, artisan de Bruijn graph… these are rare and easily startled. Be sure to approach with caution and respect its personal space. What a beautiful combinatoric flare she conveys; evolved to attract other dBgs.

🚀Excited to announce: Open-source AlphaFold3 implementation! 🚀 I am thrilled to announce one of the models we have been building for the last 8-weeks at Ligo - an open-source implementation of DeepMind’s frontier model, AlphaFold3! Here’s what we have learned, a thread (1/11):

We built a fast and memory efficient implementation in PyTorch, all under an Apache 2.0 license. In the process of building this model, we uncovered several discrepancies of the pseudocode with the deep learning literature. These are listed below. (3/11)

I recently rewrote SciCommander (tool to track #provenance of ad-hoc shell commands) in #golang while implementing a much improved approach for detecting input & output files based on great feedback I got earlier. A few words & updated poster about it at: livesys.se/posts/rewrite-…

When the actual code of a method is kept hidden (Alpha Fold 3), errors in the description and explaination *are* errors in the method, since those are the only things the authors have made transparently available to the community to analyze and understand.

For the evening crowd: PLANES: Plausibility Analysis of Epidemiological Signals 📈🦠 Blog post 📝 blog.stephenturner.us/p/planes-plaus… Preprint 📄 medrxiv.org/content/10.110… R Package #rstats📦 cran.r-project.org/package=rplanes Source code 🧬🖥️ github.com/signaturescien…

Preprint on "BWT construction and search at the terabase scale". We can compress 100 human genomes to 11GB in 21 hours, find SMEMs with it, do affine-gap alignment and retrieve similar local haplotypes. 7.3Tb commonly sequenced bacterial genomes ⇒ 30GB arxiv.org/abs/2409.00613

Excited to attend #MLCB2024 in person for the first time and present our work on understanding the resistance mechanisms of CDK4/6 inhibitors. nature.com/articles/s4301…

I'm sad to have missed WABI 24. It's one of my favorite meetings; always full of exciting work. But, I'm thrilled to be able to share publicly now that WABI 2025 will be held at Univ. of Maryland (cc UMIACS, CBCB at UMD, UMD Department of Computer Science)! We look forward to welcoming you! youtu.be/vcbGJ_ThWN0?si…

How you know you're famous for something 𝕐 (@[email protected])

I have released v0.5.0 of buttery-eel for nanopore basecalling with slow5 files github.com/Psy-Fer/butter…

JHU Computer Science is hiring in all areas (including genomics/comp bio)! Come join us! apply.interfolio.com/153420 JHU Computer Science

The source code of 10x Genomics' Cell Ranger tool suite to view, edit, modify, build upon, and re-distribute (at the very least for noncommercial purposes).

An interesting report; though I’ve not explored the methodology deeply yet. It does reinforce the (perhaps obvious) point that state schools’ strong 1st amendment requirements, as repeatedly adjudicated, serve as somewhat of a bulwark against free-speech violations.

If you work with #singlecellRNA/#scRNAseq/#singlecell data, have you ever done an analysis where you started directly from a count matrix you obtained online? Did you know the full provenance of this matrix (upstream processing, versions, options, etc.)?

"source-available" != "open-source"

Interesting observation! I guess counting sort is a "sort" in the traditional sense, but not most comparison-based algorithms.