Weekly recap (Oct 24, 2025)

Polyglot data science pipeline (R+Python+Julia) with rixpress, EuroBioC2025/BCN, Python packaging for R devs, Python in 2025, R/Pharma, DARPA, ASHG, the R Data Scientist, Quarto, universities and AI

Bruno Rodrigues: A Polyglot Data Science Pipeline with rixpress. This post demonstrates the power of the rixpress R package to build, orchestrate, and execute a fully reproducible, polyglot data science pipeline. The pipeline (1) simulates synthetic economic data using a macroeconomic model in Julia; (2) trains a machine learning model on this data to make forecasts in Python; (3) visualizes the model’s performance in R.¹

The 2025 The European Bioconductor Conference recently took place in Barcelona, one of my favorite cities in the world. Laurah Ondari writes a EuroBioC2025 conference recap on the Bioconductor community blog.

Yohann Mansiaux: Python package development for R developers.

R/Pharma conference (free!). R/Pharma is a scientifically & industry oriented, collegial online conference focused on the use of R in the development of pharmaceuticals, with breakout in-person and APAC satellite events.

University of Virginia rejects the compact.

Smart-RBC: Interesting new program from DARPA BTO to engineer red blood cells to alter human physiology (sense, decide, act) to (1) increase exercise tolerance and (2) accelerate hemostasis.

Jumping Rivers blog: What’s new for Python in 2025?

Albert Vilella: Everything you need to know from ASHG25 (partially paywalled).

Ethan Mollick’s Opinionated Guide to Using AI Right Now.

There’s now a CRAN Task View for Anomaly Detection.

The R Data Scientist (Oct 21 2025): Community & digests, stats & stories, packages & tools, authoring & visuals, applied modeling, methods & tutorials, academic research.

(From the above newsletter) clitable: a new R package to easily print pretty tables in the terminal.

(Also from the R Data Scientist newsletter above) Quarto Wizard 1.0.0: Democratising Quarto Extension Management.

See also the Quarto Extension directory.

Nature News: Universities are embracing AI: will students get smarter or stop thinking? I think both. AI is incredibly helpful in helping me learn about complex topics (most recent: quantum computing in genomics). On the other hand, the temptation to outsource reading/writing to AI is strong, and I feel dumber as soon as I do.

rOpenSci newsletter, October 2025: community call, working, new packages, peer review, package development corner, git tricks.

Great long read by Alex Wakeman on Asimov Press: Why Nigeria Accepted GMOs. Genetically modified crops are finding a foothold in the Global South, producing some unlikely leaders in agritech.

Google Blog: Google’s Gemma AI model helps discover new potential cancer therapy pathway. Here’s the preprint published a few days prior: Scaling Large Language Models for Next-Generation Single-Cell Analysis.

Finally, a few other papers and preprints that caught my attention this week:

• Microorganisms as architects of a sustainable future

• Locityper enables targeted genotyping of complex polymorphic genes

• Redefining cellular reprogramming with advanced genomic technologies

• PRISM-G: an interpretable privacy scoring method for assessing risk in synthetic human genome data

• A high-coverage genome from a 200,000-year-old Denisovan

• Generative AI for Biosciences: Emerging Threats and Roadmap to Biosecurity

• Living Framework and Guidelines for Information Disclosure in Biosecurity

• Simple scaling laws control the genetic architectures of human complex traits

• GraphComm predicts cell cell communication using a graph based deep learning method in single cell RNA sequencing data

1. Did you know that “Jupyter” is a reference to the three core programming languages supported by Jupyter, which are Julia, Python and R (source).