iiinova — where your digital self keeps evolving

Pinar Demetci @pinard.bsky.social · 29/10/2025

Tune in for a great MIA talk on ML for regulatory genomics by @jmschreiber91.bsky.social and Gregory Andrews now! 🧪 broad.io/mia (Will also be available online on our YouTube playlist later: www.youtube.com/watch?v=sMTO...)

052

Pinar Demetci @pinard.bsky.social · 02/04/2025

🧪 Super excited to host Gokcen Eraslan and Surag Nair from Genentech at MIA this Wednesday! They will describe a new multimodal masked modeling framework for functional genomics, which unifies sequence-to-function models, language models and synthetic sequence design. x.com/MIA_at_Broad...

030

Pinar Demetci @pinard.bsky.social · 12/03/2025

If you miss the talk, don't fret, we'll later post it on your YouTube playlist, which also has an awesome recent talk by @robp.bsky.social and Pratab Singh on quantifying uncertainty in inferred transcript abundances in sequencing data, with quite a detailed primer www.youtube.com/playlist?lis...

youtube.com

Models, Inference and Algorithms Meeting - YouTube

MIA homepage: http://www.broadinstitute.org/mia Models, Inference & Algorithms (MIA) is a Broad initiative to support learning and collaboration across the i...

041

Pinar Demetci @pinard.bsky.social · 12/03/2025

Tomorrow, I am super excited to be hosting Ellen Zhong & Rishwanth Raghu from Princeton at the MIA seminar, who will discuss deep learning models to reconstruct 3D protein complexes and study biomolecular dynamics from 2D cryoEM/cryoET images 🧪https://x.com/MIA_at_Broad/status/1899195133201170569

162

Pinar Demetci @pinard.bsky.social · 27/02/2025

Very interesting. So oni isn’t a nickname or something but they just use it as a completely new pronoun? This is new to me. Do you know if “oni” is meant to be for a specific gender?

110

Pinar Demetci @pinard.bsky.social · 27/02/2025

I’m Turkish and I have no idea what you’re describing here. Can you use it in a sentence as an example?

110

Pinar Demetci @pinard.bsky.social · 19/01/2025

I believe Orr has actually reached out for an MIA talk :) I will check out the previous one on YouTube; thank you for pointing me there

110

Pinar Demetci @pinard.bsky.social · 18/01/2025

Hi Liz! This sounds very interesting; do you know if the Zoom portion is only open for UPenn participants? I’m very curious to learn about what you guys have been developing so I’m wondering if there are any plans to opening the talk up to a broader audience

120

Pinar Demetci @pinard.bsky.social · 17/12/2024

It isn’t just for women, actually, men can get the vaccine, too, and it protects them against a number of cancers as well.

150

Reposted by Pinar Demetci

Brandon Amos @bdamos.bsky.social · 05/12/2024

hi everyone!! let's try this optimal transport again 🙃

232931

Pinar Demetci @pinard.bsky.social · 04/12/2024

98% appears to be an outdated number, though, no?

040

Pinar Demetci @pinard.bsky.social · 01/12/2024

should have said “attending”*

030

Pinar Demetci @pinard.bsky.social · 01/12/2024

Maybe they were paying attention to a large spectrum of time (including the upcoming couple of years) while inventing attention in 2015 /s

150

Reposted by Pinar Demetci

Ted Underwood @tedunderwood.com · 29/11/2024

I would put this even more strongly: open source AI is probably our only realistic chance to avoid a terrifying increase in concentration of power. I do not want to live in a world where the people with all the money also have all the intellectual power.

2240784

Pinar Demetci @pinard.bsky.social · 19/11/2024

I think private likes is a nice feature. I’d expect people would be split on this.

120

Pinar Demetci @pinard.bsky.social · 11/11/2024

I similarly hate reading computational papers published in Nature. I complain to my office mates every time I need to read one 😅

120

Pinar Demetci @pinard.bsky.social · 11/11/2024

I believe Nature used to mostly publish applied science, where the main message is the findings and not so much the reagents etc used to conduct an experiment. Then they started to publish in basically every field and in anything computational, methods matter a lot more to the reader.

110