Jayaram Kancherla

this game honestly feels like weekend chores...

Screenshot one of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.

Screenshot two of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.

Screenshot three of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.

chronometre: A new package (pair) demo for R and Python
Efficient yet simple object exchange between R and Python
dirk.eddelbuettel.com/blog/2026/02...
#rstats #rcpp #python #datascience

A follow up, you can fetch Organism DB objects straight from AnnotationHub! The `orgdb` Python package makes querying annotations efficient and easy.

repo: github.com/BiocPy/orgdb

#biocpy #orgdb

not sure what to do with this one

Seamlessly access R/Bioconductor annotations in Python!

The new "txdb" package lets you load standard TxDb SQLite databases and work with genomic features as native #BiocPy `GenomicRanges` objects.

Try it: `pip install txdb`
docs: github.com/BiocPy/txdb

Scikit-bio: a fundamental Python library for biological omic data analysis - Nature Methods Nature Methods - Scikit-bio: a fundamental Python library for biological omic data analysis

The scikit-bio paper in online in Nature Methods! Many thanks to our collaborators, community contributors and reviewers! We couldn’t have done it without you. www.nature.com/articles/s41... #Bioinformatics #OpenSource

The Tidyomics Ecosystem – tidyomicsBlog An introduction to the tidyomics ecosystem, including principles, core packages, publications, and community resources for tidy omic analysis.

It's official! The #tidyomics blog is up! 🚀

Here, the introductory post
tidyomics.github.io/tidyomicsBlo...

This blog aims to publish not only news about our community but also anything related to tidy interfaces, opinions, events, and more.

Plz share! 🙏
@bioconductor.bsky.social #rstats

Thanks Tommy for sharing our work. we recently updated some of our core packages in Python, especially Genomic Ranges that can perform some very fast interval operations, more here - bsky.app/profile/jkan...

GitHub - BiocPy/tutorial: tutorial for all bioconductor representations in Python tutorial for all bioconductor representations in Python - BiocPy/tutorial

github.com/BiocPy/tutor..., happy to help if you need any details

Reader Response to "AI Overinvestment" A couple of days ago, I wrote a piece titled “What if there’s an overinvestment in AI?” I received some thoughtful emails in response to that piece. One reader named Craig Melillo took the time to sh...

another good read around the same topic - www.mbi-deepdives.com/reader-respo...

If you're building Python pipelines for genomics, check out GenomicRanges. For feature requests or feedback, reach out to us on GitHub!

Try it: `pip install genomicranges`
Contribute: github.com/BiocPy/Genom...

#BiocPy

GitHub - LTLA/nclist-cpp: Nested containment lists in C++ Nested containment lists in C++. Contribute to LTLA/nclist-cpp development by creating an account on GitHub.

Now with nclist, it performs fast, complex overlap queries, all while keeping the same interface as the R equivalent. Check out Aaron's nclist @ github.com/LTLA/nclist-...

GitHub - BiocPy/GenomicRanges: Container class to represent genomic locations and support genomic analysis Container class to represent genomic locations and support genomic analysis - BiocPy/GenomicRanges

The GenomicRanges (github.com/BiocPy/GenomicRanges) package for Python got a major speed boost (v0.7.0+) thanks to Aaron’s blazing-fast C++ nclist interval indexing!

#Bioinformatics #Genomics #Bioconductor

while I agree on the h5 format, my last 5 years working with h5ad has not been easy. there's a lot of language assumptions baked in. hoping anndataR will be more stable soon

A solution I've been thinking of or may be I am unaware of is a community of developers brought together across rust, python and R. Bioconductor is organizing a rust workshop soon and seems like more of this should be happenning to bring everyone together. thoughts?

Most end users of bioinformatics tools work in R or Python, which creates a language barrier and overhead for users who want to contribute back or for newcomers getting started. Second, while I do know many awesome Rust packages, some of them often lack the user base because of limited interfaces.

h5ad/AnnData have similar problems with how its implemented in Python. it's not really language agnostic.

Kuzu - Graph Database Documentation for Kuzu, an open source, embedded graph database

not ragnar, but I've been meaning to try out the llm extension in kuzu for something similar, might be worth to check this out - docs.kuzudb.com/extensions/l...

It feels like all the AI companies are getting caught up in the hype. soon they'll realize overemphasizing prompt engineering has led to a decline in deeper technical capabilities and probably hurts them in the long run. They'll probably find a way to spin a different narrative by then...

Aaron tried switching over to the experimental 64 bit support for Kana at the beginning of the year, and I believe he mentioned there were too many issues at the time.

BiocPy has been open sourced for a while now and we continue to develop new functionality. Currently our focus has been around spatial and ml interop with bioc data structures.

while uv is great, I had issues using it inside vpn or installing from private registries.

‽

i watched the anime a few years ago and it was awesome. Didn't know they have a live action now..

Hey #genomics, #bioinformatics & #algorithms peeps 💻🧬. If you haven't seen the CfP for WABI '25 yet, check out the website wabiconf.github.io/2025/. It will be held at UMD @umdscience.bsky.social with Broňa Brejová & myself as co-chairs! Submit your exciting & late-breaking algorithmic work to WABI

Theres quite a few examples demonstrating scientific uses of webassembly, starting with our own application for single-cell analysis - www.kanaverse.org/kana/, and many others like biowasm.com, viralwasm, onnx's js runtime etc ..

GitHub - kuzudb/kuzu: Embeddable property graph database management system built for query speed and scalability. Implements Cypher. Embeddable property graph database management system built for query speed and scalability. Implements Cypher. - kuzudb/kuzu

combined with kuzu (github.com/kuzudb/kuzu), it's all I use for most of my work

I literally just installed it after reading their 1.0 release and found it very slick and minimal. I don't use many of the features from wezterm but curious to see other replies. Also simple key-value config is easy to setup.

If this generation is Gen Alpha, I say we call the next one Gen AI—because they'll probably ask ChatGPT than reddit for advice.

They need a new investor pitch now that the AI boom is winding down ...