Michael Peter Dunne

[Gina El Nesr @ginaelnesr]

since (AI-generated) protein folding (videos) is solved, we now tackle protein conformational changes

[Aya Abdelsalam Ismail @asalam_91]

[1/n] Does AlphaFold3 "know" biophysics and the physics of protein folding? Are protein language models (pLMs) learning coevolutionary patterns? You can try to guess the answer to these questions using mechanistic interpretability. But the thing is, more often than not, we know beforehand what we want/expect our model to learn, especially in very well-studied domains like Biology. So, instead of playing the guessing game, we trained a model that explicitly learns different biophysical and biochemical concepts that biologists care about.

In our concept bottleneck protein language model paper, we show that we can train the model with billions of parameters, with interpretability constraints, without performance degradation.🧵

Joint work with @nc_frey @tuomasoi @amywang01 @juliusadml @samuel_stanton_ @taylor_joren @0x00B1 @hcorrada @kchonyc and Joseph Kleinhenz at @PrescientDesign

[Michael Peter Dunne @mpdmpdmpdmpd]

@DanielRigden Nice! We noticed some similar behaviour, would be great to discuss!

[Michael Peter Dunne @mpdmpdmpdmpd]

Lovely

[Itai Yanai @ItaiYanai]

Doing good science is 90% finding a science buddy to constantly talk to about the project.

[The Nobel Prize @NobelPrize]

This year’s chemistry laureates Demis Hassabis and John Jumper have developed an AI model, AlphaFold2, to solve a 50-year-old problem: predicting proteins’ complex structures.

Check out two examples of protein structures determined using AlphaFold2. First up, a bacterial enzyme that causes antibiotic resistance. The structure is important for discovering ways of preventing antibiotic resistance.

Animation: ©Terezia Kovalova/The Royal Swedish Academy of Sciences

#NobelPrize

[Michael Peter Dunne @mpdmpdmpdmpd]

@guardiannews Or crack down on / double the postage costs for junk, circulars, unnecessary bits of paper from the bank and bills companies? Almost 100% of the mail I get goes straight in the bin.

[Michael Peter Dunne @mpdmpdmpdmpd]

4/ Here’s a StackOverflow post with a MWE https://t.co/utvb9XaJgQ!

[Michael Peter Dunne @mpdmpdmpdmpd]

1/ Hello @ONNX and @PyTorch and also X! Would you like to help me solve a mystery?

[Michael Peter Dunne @mpdmpdmpdmpd]

3/ Two colleagues can’t reproduce. It's weird and confusing and I'm scared. I’m using an @nvidia Tesla T4 with Cuda 12.0 on @awscloud. Not sure where I would file a bug or if it even is a bug and not just a weird mess!

[Michael Peter Dunne @mpdmpdmpdmpd]

@elonmusk Thank u

[Michael Peter Dunne @mpdmpdmpdmpd]

@SnoopDogg Stay strong xxx

[Michael Peter Dunne @mpdmpdmpdmpd]

These truly are the end times.

RIP x

[BBC News (World) @BBCWorld]

Nestle axes Caramac bar after 64 years https://t.co/rJcUdC4slC

[Michael Peter Dunne @mpdmpdmpdmpd]

@Steve__Kelly Uh oh

[Michael Peter Dunne @mpdmpdmpdmpd]

Wordle 761 2/6*

🟩⬛🟩🟩🟩
🟩🟩🟩🟩🟩

So close!

[Leonardo Álvarez @alvarezbiotec]

February Moon 🌝 from #Santiago #Chile.

First #Astrophoto ever.

#moon #Astrophotography

[Michael Peter Dunne @mpdmpdmpdmpd]

@nicholas_pearce This has legitimately blown my mind.

[Jo O’Meara @joomeara]

Good morning!! ❤️❤️❤️

[Michael Peter Dunne @mpdmpdmpdmpd]

@ncarleson @PhilippBayer @actually_alan Hi Nick! Good idea. Alan probably needs an update.

In the meantime you can use Alanis Morissette https://t.co/zajJ1QAkES, which is a variant of the original Alan one-liner.

[Michael Peter Dunne @mpdmpdmpdmpd]

Thanks @carlyraejepsen for spending 2,713 minutes with me in 2022. I couldn’t stop listening to Cut To The Feeling. #SpotifyWrapped https://t.co/lNWzP1600C