An iguanaβs upper body photographed up close. Behind it, a blurred second iguana can be seen
And one more, there was an iguana party on Key West this morning π¦πΏ
08.11.2025 07:04 β π 98 π 7 π¬ 0 π 0@timhowes.bsky.social
computational biologist and data scientist San Francisco Bay Area biology, evolution, statistics, open source software
An iguanaβs upper body photographed up close. Behind it, a blurred second iguana can be seen
And one more, there was an iguana party on Key West this morning π¦πΏ
08.11.2025 07:04 β π 98 π 7 π¬ 0 π 0An illustration of two alligator gar on a teal background. The fish have a long alligator-like snout and are covered in armored scales. The fish are greenish-brown and have some speckled darker spots and markings on their fins and back.
gartastic
11.09.2025 18:25 β π 279 π 76 π¬ 3 π 2'We hypothesized that cDC suppression in PDAC may further impair tumor immunity. We found that low tissue expression of Flt3L partially underlies cDC deficits. Treatment with systemic Flt3L & CD40 agonists restored cDC number & function in murine models & clinical trial samples from PDAC patients'
03.09.2025 23:32 β π 5 π 2 π¬ 1 π 0Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities Abstract Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as βcounterfactual prediction machines,β which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).
Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.
A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals). Illustrated are 1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals 2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and 3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.
Ever stared at a table of regression coefficients & wondered what you're doing with your life?
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
DuckDB may have an adorable mascot, but this is serious DB tech! As a larger-than-memory system, data durability is critical!
10.08.2025 15:03 β π 4 π 1 π¬ 0 π 0Where first graders watch educational media, before the impact of eliminating federal funds:
* PBS - 45%
* Nickelodeon - 14%
* YouTube - 9%
(from a 2025 study by child development researchers)
Galaxy brain meme format outlining options for working with data: CSV file, parquet & duckdb, databases, and remote SSH sessions
I'm excited to speak this afternoon at #useR2025 on outgrowing your laptop with #Positron for #rstats users!
You can check out my slides at juliasilge.github.io/useR-2025/
Two giant green anemones in a tidepool: they are large and circular with an olive green stalk and bright lime green tentacles fading to a more khaki color.
A couple of giant green anemones (Anthopleura xanthogrammica) in a Santa Cruz #tidepool, dwarfing a field of their smaller relative, aggregating anemones (Anthopleura elegantissima). π¦π¦ #norcal
04.08.2025 17:06 β π 19 π 4 π¬ 0 π 0Glow-in-the-dark marsupial shows off its luminous fur β Julyβs best science images
04.08.2025 17:28 β π 140 π 49 π¬ 1 π 13Had a great time on @KQED Forum.
www.kqed.org/forum/201010...
Psychologists & related fields: what is a big methods question that you believe remains somewhat unanswered & that you'd love to see adressed? There may be an opportunity for a large-scale meta-science project with many teams trying to figure out a question.
Any thoughts welcome! #psychscisky
Many attendees arrived by the Lady Rose Ferry on Sunday, sailing three and a half hours down the Alberni inlet fiord. Hereβs the ferry arriving at Bamfieldβ¦
28.07.2025 16:43 β π 0 π 1 π¬ 1 π 0Positron is definitely visually more than RStudio, and this is a helpful overview.
28.07.2025 16:57 β π 8 π 1 π¬ 1 π 0"My First R Package"
Next week Iβll be giving a hands-on workshop on #RStats We'll start from scratch and build our very own π¦
Re-use go-to functions, boost reproducibility, and share code with colleagues.
I always a blast teaching this and am excited to meet the new group!
tinyurl.com/mv2hrd8n
At this point, I might as well --
Here's an infographic showing different ways to include age as a predictor. The top shows two extremes, just as a plain old numerical predictor (imposes linear trajectory) vs. categorical predictor (imposes nothing whatsoever). And then three solutions in between!
An illustration featuring a leopard shark, bat ray, and the California state marine fish: the garibaldi. They are swimming next to giant kelp and illustrated on a vibrant blue background.
A lil tribute to the kelp forests along the California coast πβ¨
08.07.2025 17:11 β π 5731 π 808 π¬ 68 π 25A graphic-style illustration of 3 sea slugs (known as nudibranchs). Dark blue bold rounded text below the slugs reads βNudibranch Admirerβ. The slugs are vibrantly colored with hues of blue, yellow, orange, and purple and are illustrated on a saturated light blue background.
I just think sea slugs are neat π
19.06.2025 19:22 β π 1326 π 301 π¬ 36 π 7sometimes Louie will babble nonsense that vaguely sounds like @bobservo.bsky.social
12.06.2025 20:28 β π 142 π 10 π¬ 9 π 1Anglerfish ancestors once roamed the seafloor. Here's how we know : Short Wave : NPR
05.06.2025 20:16 β π 32 π 12 π¬ 0 π 1The super-talented artist and author Rosemary Mosco is donating proceeds from her shop in the month of June to IPB! There's an image of a poster of female North American birds, a yellow warbler tshirt, and a mug that says Mornings Can Kiss My Cloaca.
All proceeds from my @topa.to store in June go to the Institute for Bird Populations, @instbirdpop.bsky.social! Times are tough for nonprofits, and monitoring bird populations is more important than ever. Show your support with a tasteful print or tasteless mug. topatoco.com/collections/...
01.06.2025 13:43 β π 164 π 61 π¬ 3 π 1A male Baltimore Oriole flying towards his next tree top to perch and sing his clear and sweet song.
27.05.2025 22:19 β π 307 π 30 π¬ 3 π 1#DuckDB DuckLake β a new lakehouse format built on pure SQL β is out!
duckdb.org/2025/05/27/d...
duckdb.org/docs/stable/...
In the US at least, you can get the full series through Amazon Prime, if you get the PBS add-on subscription. Also includes things like Mr. Rogers and Kratts' Creatures.
26.05.2025 19:38 β π 1 π 0 π¬ 0 π 0I would say that true case-control studies are not that common in epi any more, IME odds ratios get used in clinical research mostly because you get them by pressing the "logistic regression" button and doing zero additional thought. My brain dump on ORs here: cameronpatrick.com/post/2023/07...
26.05.2025 13:30 β π 14 π 4 π¬ 2 π 0Goodnight, sun
24.05.2025 03:08 β π 8 π 1 π¬ 0 π 0New paper from the lab: Our teeth arose as sensory organs on the outside of the body of ancient jawless fish.!! Congrats to Yara Haridy and the team!
Background and video: phys.org/news/2025-05...
Open Access Paper: www.nature.com/articles/s41...
News and Views: www.nature.com/articles/d41...
βExploratory data analysisβ and βconfirmatory data analysisβ are the same thing.
statmodeling.stat.columbia.edu/2025/05/23/e...
Pretty little babies π₯°πβ
β
The pearl octopus (Muusoctopus robustus) mates and nests in deep-sea thermal springs, where the warm water seeping from the seafloor accelerates the development of their embryos.β
Learn more: www.mbari.org/animal/pearl...
Same energy π
23.05.2025 21:20 β π 74 π 11 π¬ 0 π 0