ElnaggarLab/ankh3-xl · Hugging Face We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Ankh3 Large weights: huggingface.co/ElnaggarLab/...
Ankh3 XL weights: huggingface.co/ElnaggarLab/...

Beyond Simple Concatenation: Fairly Assessing PLM Architectures for Multi-Chain Protein-Protein Interactions Prediction Protein-protein interactions (PPIs) are fundamental to numerous cellular processes, and their characterization is vital for understanding disease mechanisms and guiding drug discovery. While protein l...

Check out our latest research on sequence level protein-protein interaction.

We evaluated multiple different protein language models with different architectural strategies for adapting PLMs to multi-chain PPI prediction.

Arxiv: arxiv.org/abs/2505.20036
Repo: github.com/Proteinea/pp...

Ankh3: Multi-Task Pretraining with Sequence Denoising and Completion Enhances Protein Representations Protein language models (PLMs) have emerged as powerful tools to detect complex patterns of protein sequences. However, the capability of PLMs to fully capture information on protein sequences might b...

Check out Ankh V3, our latest research on protein language models. Ankh V3 is a protein language model powered by the T5 architecture and UL2 objective.

Arxiv: arxiv.org/abs/2505.20052

Are there machine learning folks who are working on protein/dna language models that I should follow?

Just discovered bluesky couple of days ago and I found many of my favorite researchers are way more active here instead of twitter.