iiinova — where your digital self keeps evolving

Reposted by Shubhra Agrawal

Žiga Avsec @avsecz.bsky.social · 25/06/2025

Excited to launch our AlphaGenome API goo.gle/3ZPUeFX along with the preprint goo.gle/45AkUyc describing and evaluating our latest DNA sequence model powering the API. Looking forward to seeing how scientists use it! @googledeepmind

521982

Reposted by Shubhra Agrawal

Nils Gilman @nilsgilman.bsky.social · 17/06/2025

Saying LLMs are worthless/pernicious because they sometimes make mistakes & is causing a general decline in reading skills (true) is like saying Google Maps is worthless/pernicous just because it sometimes gives you bad directions & has caused a general decline in human navigation skills (also true)

6376

Shubhra Agrawal @shubhra-a.bsky.social · 21/01/2025

in the meantime, that's a cool hackathon idea!

010

Shubhra Agrawal @shubhra-a.bsky.social · 21/01/2025

TIL 10/365: blood genotyping is not as straightforward as most of us have been taught. the most surprising part for me was that some people can safely take blood from 1 type but can't donate to the same type. this TIL is inspired by the thread here -> bsky.app/profile/alan...

000

Shubhra Agrawal @shubhra-a.bsky.social · 21/01/2025

TIL 9/365: Boehringer Ingelheim has an open science portal opnMe where they pose unsolved challenges in different therapeutics areas and invite participants to share their solutions. Promising ideas get funding support! Check out the list of challenges: www.opnme.com/opn2experts

opnme.com

010

Shubhra Agrawal @shubhra-a.bsky.social · 21/01/2025

TIL 8/365: Phone SIM cards have a shelf life 😆 Sounds obvious in retrospect, never considered it as a possible explanation for my near constant network failure. ran around customer care portals for a month before someone suggested i change it 🫠

000

Shubhra Agrawal @shubhra-a.bsky.social · 21/01/2025

TIL 7/365: Being a 'woman' is hard i decided to learn the basics of styling and makeup this year. but beyond the active time spent shopping/trying products on, there is so much homework that the average woman does on fabrics,design,skin types,hair types,markets and trends. i feel like a toddler 🧒

010

Reposted by Shubhra Agrawal

Mark O. Martin @markowenmartin.bsky.social · 21/01/2025

Here goes—-on with my day!

0416

Shubhra Agrawal @shubhra-a.bsky.social · 17/01/2025

TIL 6/365: US is now going to control the export of cell sorters and mass spec machines because biological data + AI poses a security risk in the wrong hands. huh. i'm not familiar with wet-lab, unfortunately. it might make large scale NGS harder in the short term but i'm sure there are ways ??

000

Reposted by Shubhra Agrawal

Simon P. Couch @simonpcouch.com · 16/01/2025

Meet gander, a chat interface for R data scientists🪿 gander automatically incorporates context from your #rstats environment and surrounding code and can be used with any model supported by ellmer in either RStudio or Positron. Read more: simonpcouch.github.io/gander/

613838

Shubhra Agrawal @shubhra-a.bsky.social · 17/01/2025

TIL 5/365: My ADHD ass is dysregulated 90% of the time. I just never realised it before because I associate emotional dysregulation with bursts of temper, or extreme mood swings (: apparently, flip-flopping between internal screaming and static-noise-no-thought phases counts too.

010

Shubhra Agrawal @shubhra-a.bsky.social · 15/01/2025

TIL 4/365: I am helpless against the Lego shelves 😩 Especially the 3-in-1 kind

000

Reposted by Shubhra Agrawal

Mike Feigin 🥯 @mikefeigin.bsky.social · 14/01/2025

Please for the love of Satan, post more jokes!!

041

Shubhra Agrawal @shubhra-a.bsky.social · 14/01/2025

TIL 3/365: there is an upper limit to how many people you can directly manage while working on active projects yourself :| i also think I blew wayy past this limit months ago. any sage advice from people who've been there? 😅

000

Reposted by Shubhra Agrawal

Shubhra Agrawal @shubhra-a.bsky.social · 12/01/2025

One of my focus areas for 2025 is to 'build publicly' at work. Not a novel concept, but definitely an area of improvement for me. but why stop there? so i'll be starting a TIL (Today, I learned) series on bsky with the aim to share 365 TIL posts this year.

media.tenor.com

a man with his mouth open and the words one after the other behind him

Alt: gif of David from Schitts creek saying "one after the other, consecutively, in a row"

111

Reposted by Shubhra Agrawal

Hani Goodarzi @genophoria.bsky.social · 13/01/2025

With 100 millions single cell gene expression measurements, using >1000 drugs across 60 cell line models, Tahoe-100M is larger than all public single-cell transcriptomic data combined. And it is 50x larger than all drug-perturbed single-cell data put together.

131

Shubhra Agrawal @shubhra-a.bsky.social · 13/01/2025

TIL 2/365: core facilities in academia have a transparency problem researchers get access to the raw data, and the processed excel file from the core. The in-between details important for drawing any verifiable conclusions? gone. even basics like the target library used for LCMS could be missing 🤷‍♀️

010

Shubhra Agrawal @shubhra-a.bsky.social · 12/01/2025

TIL 1/365 lncRNAs might be a lot more important for single cell: cell type annotation than I previously thought (okay, I never thought about it before 🤷‍♀️) most of the current public SC datasets don't even capture lncRNAs consistently. def going to explore this more

000

Shubhra Agrawal @shubhra-a.bsky.social · 12/01/2025

media.tenor.com

a man with his mouth open and the words one after the other behind him

Alt: gif of David from Schitts creek saying "one after the other, consecutively, in a row"

111

Reposted by Shubhra Agrawal

Ariel Rokem @arokem.org · 10/01/2025

Quarto is great! Another option that elegantly addresses this issue is jupytext: jupytext.readthedocs.io/en/latest/, which creates a version-control friendly twin of your notebook and keeps them in sync as you work.

jupytext.readthedocs.io

Jupytext — Jupytext documentation

121

Reposted by Shubhra Agrawal

Bo Wang @bowang87.bsky.social · 10/01/2025

How to build a virtual cell is the “Holy Grail” of the biology! A great read from The Atlantic @theatlantic.com ! Also thanks for highlighting our scGPT, the model that started the new wave of foundation models in biology! www.theatlantic.com/technology/a...

theatlantic.com

A Virtual Cell Is a ‘Holy Grail’ of Science. It’s Getting Closer.

Large language models may unlock a new and valuable type of research.

0114

Reposted by Shubhra Agrawal

Kieran Campbell @kieranrcampbell.bsky.social · 10/01/2025

When asked about knockout vs knockdown, Google AI wonderfully bet-hedges on whether you're interested in genetics or boxing

3165

Reposted by Shubhra Agrawal

Richard Sever @richardsever.bsky.social · 07/01/2025

Btw - this is another case where my frequent point that not just IF but also PubMed/PMC are obstacles to innovation applies: if a journal does a novel form of post-hoc change that doesn't fit in PubMed's rigid framework, it's not registered (so is disincentivized)

293

Reposted by Shubhra Agrawal

Ted Underwood @tedunderwood.com · 05/01/2025

Part of the reason people are glum about AI is that researchers themselves represent the goal as full equivalence for human agency (“AGI”). While I get the theoretical interest, I think what most customers want (and what we’ve really produced so far) is closer to “prosthetics for thinking.”

2219627

Reposted by Shubhra Agrawal

Kieran Campbell @kieranrcampbell.bsky.social · 01/01/2025

Hopeful it’s the year use cases that add value and/or improve outcomes actually crystallize. Seem to be stuck between lots of hype on one end and people convinced of 0 utility on the other

051

Reposted by Shubhra Agrawal

Stats @final-stats.bsky.social · 30/12/2024

This is absolutely amazing news. Currently it takes an average of 10 YEARS for women in the US to be diagnosed with endometriosis. A decade of agony, of horrible painful suffering every month could be eliminated for millions of women.

211737

Reposted by Shubhra Agrawal

Robert Manchild @robertmanchild.com · 27/12/2024

What’s it called when you can’t get motivated and also can’t relax because I’m that.

2418432

Shubhra Agrawal @shubhra-a.bsky.social · 27/12/2024

The martian scientists were evaluating the conditions on earth based on their own understanding of life-sustaining qualities. earth's gravity disqualified any possibility of larger life forms. earth's oxygen meant there would be uncontrolled fires. funny how biases work :)

000

Shubhra Agrawal @shubhra-a.bsky.social · 27/12/2024

Lovely read. Some obscure childhood stories disproportionately inform our current understanding of the world. I'm reminded of a short story I read in high school; it was a scientific record from Mars evaluating the earth as a livable planet. spoiler alert: Earth Failed!

110

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

All I'm saying is, you can't turn back the clock. Once a piece of tech is out, good or bad, it's here to stay. We can (and should) push for more moderation, regulation, etc while you arm your students with the best practices and critical thinking skills to deal with it. Not fishing for a response✌️

000

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

But the examples you're giving? Students will ALWAYS try and find ways to reduce the effort. Google search/wikipedia/quora was my generation's shortcut. But our teachers pointed out the holes and we learned how to balance speed with depth. Fact check wiki info etc. It'll happen with AI tools too 🤞

100

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

Bad phrasing, I meant the AI agents are also pretty idiotic a lot of the times. Not the concerns lol

000

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

And policy makers who accept that this cat (idiot AI agent) is out of the bag now. Tantrums and fear mongering would do nothing except push your students away. The internet is a lying machine, social media is a lying machine. Arming students with critical thinking skills? Priceless

100

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

There are legitimate concerns around how learning methods will need to be adapted in the age of freely available "AI agents" that are also pretty idiotic. And how fast. Not trying to belittle any of that. But those changes will always be driven by educators (contd.)

210

Shubhra Agrawal @shubhra-a.bsky.social · 25/12/2024

It's part of the natural discourse cycle any time a truly transformative technology becomes popular. Love the hashtag though

020

Reposted by Shubhra Agrawal

Raphael Gottardo @rgottardo.bsky.social · 22/12/2024

Precision Dermatology 2.0: #DataScience + #Immunology + #Genomics + #Clinical Science. Interdisciplinary collaboration at its best at the @CHUV. www.chuv.ch/fr/chuv-home... Paper on @naturecomms.bsky.social www.nature.com/articles/s41...

chuv.ch

Eczéma, psoriasis, lupus et autres maladies inflammatoires de la peau : des signatures génomiques pour améliorer le diagnostic et le traitement

0112

Reposted by Shubhra Agrawal

BrunetLab @brunetlab.bsky.social · 18/12/2024

Thrilled to see this paper out!! 🧪 Spatial transcriptomics of brain aging and 'spatial aging clocks' identify cells that have pro-aging or pro-rejuvenating effects on their neighbors! Huge CONGRATS to Eric Sun and all authors! Fantastic collaboration with @jameszou.bsky.social! rdcu.be/d33uQ 🧵

rdcu.be

Spatial transcriptomic clocks reveal cell proximity effects in brain ageing

Nature - A spatially resolved single-cell transcriptomics map of the mouse brain at different ages reveals signatures of ageing, rejuvenation and disease, including ageing effects associated with T...

2230296

Shubhra Agrawal @shubhra-a.bsky.social · 16/12/2024

burn-out induced fermentation weekend is over, back on the corporate saddle for the week 🫡 Send prayers

000

Shubhra Agrawal @shubhra-a.bsky.social · 11/12/2024

That's hard to prevent, people have been doing the same with stackoverflow responses for years. The difference is that students might trust AI responses more than crowd responses on websites like stackoverflow. It's important to showcase that these models are not infallible

110

Shubhra Agrawal @shubhra-a.bsky.social · 11/12/2024

I keep hearing that coding assistance in IDEs (copilot included) can be very annoying, haven't tried it myself yet.

020

Shubhra Agrawal @shubhra-a.bsky.social · 11/12/2024

That's a great point. I haven't tried asking for full solutions yet, with any model so I can't comment on that. But chatgpt definitely goes in a loop after a few iterations, wonder why. It's supposed to retain the session memory

000

Reposted by Shubhra Agrawal

Timothy Stiles @stiles.io · 10/12/2024

I love when people post the tiny little details of their lab work here.

082

Shubhra Agrawal @shubhra-a.bsky.social · 11/12/2024

Some learnings from my experiments with chatgpt and Claude: - great at reviewing code or error messages if given proper context - great for working with unfamiliar analysis packages, troubleshooting (but in a prog language you understand well) - Claude >> chatgpt for bioinformatics

110

Shubhra Agrawal @shubhra-a.bsky.social · 11/12/2024

🖥️🧬 folks, when using chatbots, what are some #Bioinformatics coding problems you've had good or bad experiences with? Care to comment?

220

Shubhra Agrawal @shubhra-a.bsky.social · 10/12/2024

It's time to lean into the idea of using chatbots as coding assistants! imo focus should be on how to arrive at a reasonable solution: when to ask for tooling suggestions, vs when to ask for approaches vs when to ask for code snippets etc. there's absolutely a place for this early on!

010