iiinova — where your digital self keeps evolving

Reposted by Amy Lu

Milot Mirdita @milot.bsky.social · 20/01/2026

My time in @martinsteinegger.bsky.social's group is ending, but I’m staying in Korea to build a lab at Sungkyunkwan University School of Medicine. If you or someone you know is interested in molecular machine learning and open-source bioinformatics, please reach out. I am hiring! mirdita.org

mirdita.org

Mirdita Lab - Laboratory for Computational Biology & Molecular Machine Learning

Mirdita Lab builds scalable bioinformatics methods.

710756

Reposted by Amy Lu

Martin Steinegger 🇺🇦 @martinsteinegger.bsky.social · 26/01/2025

Just coincidentally found GenBank Release 84.0 from 1994 in the neighboring lab. Anyone out there with an even older version?

68013

Amy Lu @amyxlu.bsky.social · 28/12/2024

In case you missed our ML for proteins seminar on CHEAP compression for protein embeddings back in October, here it is -- thanks @megthescientist.bsky.social for doing so much for the MLxProteins community 🫶

2161

Reposted by Amy Lu

Meg T (she/her/hers) @megthescientist.bsky.social · 16/12/2024

•introduced “zero shot prediction” as a question of guessing a bioassay’s outcome by likelihoods of pLMs •commented on biases in evolutionary signals from Tree of life used to train pLMs (a favorite paper I read in 2024: shorturl.at/fbC7g)

181

Amy Lu @amyxlu.bsky.social · 10/12/2024

Another straightforward application is generation, either by next-token sampling or MaskGIT style denoising. We made the tokenized version of CHEAP to do generation, and decided to go with diffusion on continuous embeddings instead — but I think either would’ve worked

060

Reposted by Amy Lu

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · 09/12/2024

We trained a model to co-generate protein sequence and structure by working in the ESMFold latent space, which encodes both. PLAID only requires sequences for training but generates all-atom structures! Really proud of @amyxlu.bsky.social 's effort leading this project end-to-end!

25711

Amy Lu @amyxlu.bsky.social · 06/12/2024

immensely grateful for awesome collaborators on this work: Wilson Yan, Sarah Robinson, @kevinkaichuang.bsky.social, Vladimir Gligorijevic, @kyunghyuncho.bsky.social, Rich Bonneau, Pieter Abbeel, @ncfrey.bsky.social 🫶

030

Amy Lu @amyxlu.bsky.social · 06/12/2024

6/ We'll get to share PLAID as an oral presentation at MLSB next week 🥳 In the meantime, checkout: 📄Preprint: biorxiv.org/content/10.1... 👩‍💻Code: github.com/amyxlu/plaid 🏋️Weights: huggingface.co/amyxlu/plaid... 🌐Website: amyxlu.github.io/plaid/ 🍦Server: coming soon!

biorxiv.org

161

Amy Lu @amyxlu.bsky.social · 06/12/2024

5/🚀 ...and when prompted by function, PLAID learns sequence motifs at active sites & directly outputs sidechain positions, which backbone-only methods such as RFDiffusion can't do out-of-the-box. The residues aren't directly adjacent, suggesting that the model isn't simply memorizing training data:

conditioning on organism and function shows that PLAID has learned active site residues and sidechain positions!

152

Amy Lu @amyxlu.bsky.social · 06/12/2024

4/ On unconditional generation, PLAID generates high quality and diverse structures, especially at longer sequence lengths where previous methods underperform...

150

Amy Lu @amyxlu.bsky.social · 06/12/2024

3/ I was pretty stuck until building out the CHEAP (bit.ly/cheap-proteins) autoencoders that compressed & smoothed out the latent space: interestingly, gradual noise added to the ESMFold latent space doesn't actually corrupt the sequence and structure until the final forward diffusion timesteps 🤔

noising by a diffusion schedule in the latent space doesn't always correspond to the same corruption in the sequence and structure space...

140

Amy Lu @amyxlu.bsky.social · 06/12/2024

2/💡Co-generating sequence and structure is hard. A key insight is that to get embeddings of the ESMFold latent space during training, we only need sequence inputs. For inference, we can sample latent embeddings & use frozen sequence/structure decoders to get all-atom structure:

140

Amy Lu @amyxlu.bsky.social · 06/12/2024

1/🧬 Excited to share PLAID, our new approach for co-generating sequence and all-atom protein structures by sampling from the latent space of ESMFold. This requires only sequences during training, which unlocks more data and annotations: bit.ly/plaid-proteins 🧵

112437