python vs R?
base R vs tidyverse?
framework X vs Y?
unless you're getting paid, use the one that brings you the most joy.
life's too short to keep on writing code that you neither enjoy nor get paid for.
#rstats
I always do this in intro classes. Setting vs. Mapping in ggplot2 is so confusing untill you get it. So I try to always expose it to beginners early
A minimalist isometric 3D illustration on a dark background showing a central glowing cube labeled ".dotfiles" with a git icon. Glowing neon lines radiate from the cube to icons representing "Shell Config," "Editor & Tooling," and "AI Agents (Claude & OpenCode)." A sidebar labeled "Skills Marketplace" displays specialized icons for R, Shiny, Quarto, and writing voices, illustrating a unified development environment.
My dotfiles now manage my AI coding agents 🤖 Same instructions for Claude Code and OpenCode, symlinked from one repo. Here's how I set it up: https://drmo.site/CCrf37
neuromapr 0.2.1 has been accepted and published on CRAN!
Very excited to get this out to users in the simplest way possible, and hope the #rstats #neuroimaging community finds it useful!
lcbc-uio.github.io/neuromapr/
Episode 222 of the R Weekly Highlights Podcast is out! serve.podhome.fm/episodepage/...
🤔 Choosing LLMs with vitals @smachlis.bsky.social
👀 Pick a license (VP Nagraj)
📍 Git commits, mark your stitches @masalmon.eu
h/t @mike-thomas.bsky.social & Sam Parmar 🙏
#rstats #databs
Lovely first day of the #rainbowR conference with a trivia quiz and social by @libbyheeren.bsky.social and @damiepak.bsky.social 🥰🌈
Planning to finally post about #SEM tomorrow. If you have questions about working with SEM or you use #lavaan or #OpenMx please let me know in the comments! #RStats
I personally use ggthemes and the neat wrapper package #tidyplots (tidyplots.org) to quickly get to a professional-looking plot! 🖼️ The tidyplots developer @jbengler.de has a lot of example code available on his profile.
What is your favorite #DataViz hack? 📈
#RStats #RLadies
Foto of a feminine nonbinary person in a purple floral shirt in front of a purple poster. The poster is displaying a structural equation model analysing the influence of parental education on children’s school grades. The person is smiling cheekily to the side.
Foto of a child with four fluffy buns and blue sunglasses. The child has lowered the sunglasses below their nose and is gazing over them.
Hi, I’m this week’s curator, Josi! I‘m a PhD student in statistical genetics 🧬 and use #RStats to do #SEM. I dabble in #DataViz and recently wrote my first #Rpackage. My hobbies are reading #Fantasy books 📚 and #maximalism. You may catch me at a conference matching my outfit to my poster 🤓🧜🏽♀️ #rladies
*Pipe-friendly atlas manipulation* — atlas_region_keep(), atlas_view_keep(), and friends filter, rename, and reorder regions. Output goes straight to geom_brain().
Available on CRAN and r-universe. Full writeup: ggsegverse.github.io/news/ecosyst...
#rstats #neuroscience #datavisualization
*Cortical atlases store vertex indices, not mesh geometry* — Each region is a vector of integers into a shared fsaverage5 mesh. Atlas packages shrink. Rendering loads one mesh per hemisphere instead of dozens.
*3D engine rebuilt on Three.js* — Plotly is out. Three.js through htmlwidgets gives orthographic projection, glassbrain overlays, flat shading for exact color matching, and standard anatomical camera positions.
*Tract atlases in 3D* — White matter tracts had no 3D representation in v1. They do now.
What's new:
*Unified atlas format* — A single ggseg_atlas class carries sf geometries, vertex indices, region metadata, and color palettes. Both geom_brain() and ggseg3d() read the same object.
ggseg 2.0: One atlas object for both 2D and 3D brain visualization.
In v1, you loaded `dk` for 2D and `dk_3d` for 3D.
That's gone. One call — dk() — feeds both renderers.
ggsegverse.github.io/news/ecosyst...
#rstats #neuroscience #datavisualization
I'm not sure if I'd call it routinely. But yes. And sometimes it's so bad that I need to quit positron as a whole to get things working again. But I've yet to see a pattern or anything I can reproduce.
Curator Josi, PhD candidate at Vrije Universiteit Amsterdam. Josi is currently working on a simulation study on a structural equation model using the OpenMx package.
The first #RLadies curator of 2026 is @josefina-bernardo.bsky.social 💜
I'm very grateful that Luis maintains this guide. Very much recommend to those who haven't poked around there before! #rstats
Indeed, this last step allowed us to notice from key default behaviour differences between R and python that we could adjust so the default behaviour of both implementations could align.
3. Claude could extract this information and digest it quite easily. In particular, useful when asked to setup a testing structure directly against the python implementation, so we could catch discrepancies.
This got developed in speed time, because of three factors:
1. Great source references, the reference paper outlines the intent and use-cases so well.
2. The python implementation already exists, so clear source code I could directly compare to
neuromapr generates surrogate brain maps preserving the spatial autocorrelation structure of the original. Eight null model methods --- variogram matching, Moran spectral randomization, spin tests, and more. Three lines of code to get a spatially-corrected p-value.
Standard correlation tests assume independence between brain regions. The brain violates that assumption everywhere. The result: inflated p-values and false positives in map-to-map comparisons.
Two brain maps correlate at r = 0.6. Is that real, or is it spatial autocorrelation doing what spatial autocorrelation does?
It implements the framework from Markello et al. (2022, Nature Methods) and is aligned with the neuromaps Python reference implementation. Co-developed with Claude Code.
netneurolab.github.io/neuromaps/in...
New neuromapr #rstats package is available from GitHub! Highly experimental, early adopters and bug identifiers are super welcome to report issues!
github.com/lcbc-uio/neu...
#neuroscience #rstats #neuroimaging
I'm really happy you think so!
I realy very heavily on r-universe to achieve this. Its got such a lovely API to work with (thank you @jeroenooms.bsky.social !!) so its quite low effort to write a bit of js (or have #claude help out) to keep things updated without having to render.
should I call it ggsegiverse? Maybe I should? break all the things for that change? I just might :P But sure, I can do a post on that in not too long :) I also need to finish the post I have in draft for you :P
making the website (ie. pulling from r-universe) or about the ecosystem as a whole?
We have Aal already made! github.com/ggseg/ggsegAal
Watch out for some new developments in the ecosystem over the next months! We are very close to releasing some major improvement.
