iiinova — where your digital self keeps evolving

Ryan Bartelme @microbialbart.bsky.social · 01/10/2026

I self host gitea for git repositories and kanban boards.

110

Ryan Bartelme @microbialbart.bsky.social · 28/09/2026

Is it stealing if you didn't shut off the "use my data to train your models" or didn't read the Terms and Services?

100

Ryan Bartelme @microbialbart.bsky.social · 25/09/2026

This is one of the worst starts to a Packers season in a very long time...

000

Ryan Bartelme @microbialbart.bsky.social · 25/09/2026

Vercel has Express for NodeJS, maybe that will do it?

120

Ryan Bartelme @microbialbart.bsky.social · 23/09/2026

And did an ad campaign that doesn't translate into Spanish...

030

Ryan Bartelme @microbialbart.bsky.social · 23/09/2026

Past admin (checks notes) paid McKinsey to rebrand

150

Ryan Bartelme @microbialbart.bsky.social · 23/09/2026

Hired postdoc to be assistant AI director or whatever...

010

Ryan Bartelme @microbialbart.bsky.social · 23/09/2026

Meanwhile CAIO ($400k+...wastes money flying business class) hired his unqualified postdoc who lives in LA, collects $200k+ BUT WHERE DID THE MONEY GO?!

110

Ryan Bartelme @microbialbart.bsky.social · 22/09/2026

I genuinely dislike Discord as a platform. Eats memory in the browser, eats memory in the app.

static.klipy.com

Spirited Away: No-Face Eating Cake

ALT: Spirited Away: No-Face Eating Cake

000

Ryan Bartelme @microbialbart.bsky.social · 22/09/2026

Operational security is something most people need to learn like yesterday. Don't give agents CC, non-scooed API keys, bank accounts, etc.

020

Ryan Bartelme @microbialbart.bsky.social · 21/09/2026

This is a wonderful opportunity for any marine microbiologists!!

010

Ryan Bartelme @microbialbart.bsky.social · 18/09/2026

@uncultured.carinilab.com I had a professor who pulled problems from "Consider A Spherical Cow" to warm the class up to this style of thinking

020

Ryan Bartelme @microbialbart.bsky.social · 13/09/2026

On an academic AI slack group one person pointed out this sudden "cause for alarm" around LLM's from Anthropic, OpenAI, et al. amounts to "regulatory capture". I.e. the rule will be written to enshrine their business interests while they quietly nix their IPOs citing "problems". #AI #LLMs

011

Ryan Bartelme @microbialbart.bsky.social · 12/09/2026

What floors me is that if it is indeed "so dire" then why be complicit? If they IPO your stock options will be embargoed, so what's the motivation to stay?

010

Ryan Bartelme @microbialbart.bsky.social · 12/09/2026

I spent many years at the bench before going full time computational biology/bioinformatics. It floors me how "solved" people think biology is. Even human biology is exploring the "noncoding dark genome" because we don't know diddly about squat.

030

Ryan Bartelme @microbialbart.bsky.social · 12/09/2026

"What if we hung out like a podcast without microphones." -some tech bro

011

Ryan Bartelme @microbialbart.bsky.social · 10/09/2026

I had a friend (AI infra engineer) text me about this yesterday. He sort of assumed that biology research is as automated as a Ford assembly line. Who's gonna tell them that even if you obtain information that doing the actual lab work doesn't mean it will work, work.

static.klipy.com

Gilfoyle: Honest or Nice?

ALT: Gilfoyle: Honest or Nice?

2415

Reposted by Ryan Bartelme

José Vargas-Muñiz @mycorican.bsky.social · 09/09/2026

That is a good question! The 2 drugs that we used are clinical immunosuppressants, so if they behave as they do in humans, I guess that they would also immunosuppress bats (but that is a huge assumption). Fungal & mammal calcineurin are a tad different; hopefully, it can be exploited to fight fungi

012

Ryan Bartelme @microbialbart.bsky.social · 09/09/2026

Ok weirdly did an undergrad research project on Calcineurin pathway in Blastomyces...how does this not affect bats?! I thought calcineurin inhibition was bad for all Euks

100

Ryan Bartelme @microbialbart.bsky.social · 04/09/2026

I think this is why I like using GFF3 to catch parallel annotations from multiple annotation tools

000

Ryan Bartelme @microbialbart.bsky.social · 04/09/2026

If your lessons are looking at vectors/edits/ORF's, this might get you there: www.vectorbee.com/en/

010

Ryan Bartelme @microbialbart.bsky.social · 02/09/2026

"This is more of a comment than a question..."

020

Ryan Bartelme @microbialbart.bsky.social · 29/08/2026

pandoc does this for many, many formats in the command line. 😁

010

Reposted by Ryan Bartelme

Noah Greifer @noahgreifer.bsky.social · 03/08/2026

Fascinating and clear paper by @corymccartan.com and @melodyyhuang.bsky.social, greatly enhancing our understanding of how Bayesian Additive Regression Trees (BART) works and why it is so effective. A must-read for my fellow BART enthusiasts. #statssky #causalinference

13911

Ryan Bartelme @microbialbart.bsky.social · 28/07/2026

Would love to see someone do GEM without commercial FORTRAN dependencies one of these days...Gurobi and IBM commercial licenses are so $$$.

000

Ryan Bartelme @microbialbart.bsky.social · 28/07/2026

Opt out from...Concur's "AI"? What's the worry here? The opacity of the contract? Or is it "AI = Bad"?

200

Ryan Bartelme @microbialbart.bsky.social · 28/07/2026

Bribed to do what exactly? Bolt what is effectively a chatbot with image to text analysis into a bloated SaaS? Force unis to use Concur?

100

Ryan Bartelme @microbialbart.bsky.social · 28/07/2026

Concur is awful, but the mention of "AI" is doing too much heavy lifting here. Extracting text from an image is a classical computer vision problem, even easier if it's not handwriting 🤣😂

001

Ryan Bartelme @microbialbart.bsky.social · 15/07/2026

I wonder how many of these studies are confounded by sampling effects? pubs.rsc.org/ay/article/1...

pubs.rsc.org

Submit This Form

110

Ryan Bartelme @microbialbart.bsky.social · 03/07/2026

Yep any whiff of bio and it boots you down to Opus

010

Ryan Bartelme @microbialbart.bsky.social · 03/07/2026

@cbo.bsky.social doing the heavy lifting for biologists here 🏋️

001

Ryan Bartelme @microbialbart.bsky.social · 03/07/2026

This is a good honest take, but consider that frontier models like Fable5 are blanket blocking work on any and all bioinformatics programming. This is the most tractable area for AI + biology but stands gate kept for those without the financial means of access.

010

Reposted by Ryan Bartelme

A. Murat Eren (Meren) @merenbey.bsky.social · 28/06/2026

@hifmb.de advertised 5 post-doc positions to work on social science / basic science interface, and here is a blog post to tell you why I care about this, and what it looks like: merenlab.org/blog/HIPP-on... Please spread the word if you can.

merenlab.org

Five Postdocs, Five Questions: An Experiment in Transdisciplinary Thinking at HIFMB

Some personal notes on what we are trying to achieve, and how we are planning to achieve that

01110