🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Comments welcome - our updated protein manuscript now applies the same model architecture to a clustered split "Using Deep Learning to Annotate the Protein Universe" now on BioRxiv biorxiv.org/content/10.110…

Amazing experience working with Alex Bateman and the team at EMBL-EBI to leverage neural networks for protein domain annotation!

Predicting the function of proteins using #neuralnetworks can greatly aid medical research. We are proud to collaborate with EMBL-EBI to extend the The Pfam database database of functionally annotated proteins by 10% as part of the v34.0 release. Learn more below! xfam.wordpress.com/2021/03/24/goo…

#AI is revolutionising protein science. Can it add to our knowledge of protein function? New Nature Biotechnology research shows how deep learning models can be used to improve protein annotations within The Pfam database and help predict protein function. nature.com/articles/s4158…

Using deep learning to annotate the protein universe go.nature.com/3v4tKlv

Nature Biotechnology Google AI The Francis Crick Institute MIT CSAIL EMBL-EBI Cambridge Chemistry The Pfam database Alex Bateman Lucy Colwell Thanks for all the interest everyone, and to the journal for the very helpful review. Here's a free link to the complete paper: rdcu.be/cHsqz

Here is the code and google-colab notebook if anyone wants to try! github.com/sokrypton/Cola… feedback is welcome 😀

A reminder that Chloe Hsu's talk on Learning Protein Fitness Models from Evolutionary and Experimental Data is happening on March 1! Sign up here for updates: ml4proteinengineering.com

A smooth career path for many academics. nature.com/articles/s4156… 🙏Susanne Täuber Morteza Mahmoudi, PhD #bullying #whistleblowing #corruption

Machine learning can help read the language of life: blog.google/technology/ai/… Discussing our work in Nature Biotechnology, our interactive preprint, and why Google is involved.

Our interactive manuscript preprint is available at google-research.github.io/proteinfer/

A giant thank you to all who made this work happen.

With help from #deeplearning models, The Pfam database has expanded by almost 10% exceeding all expansion made to the database over the last decade 🤯 Alex Bateman discusses this work and what the future holds for protein family classification. ebi.ac.uk/about/news/per…

Now we can google protein function google-research.github.io/proteinfer/

A new version of the proteinDB has been released on MGnify. More than double in size, it now has over 2.4 billion non-redundant sequences comprising 623 million clusters. Internal improvements mean that the DB now includes crucial links between the sequences and their metadata

MolPC is out. nature.com/articles/s4146…

Ever got a result back saying uncharacterised protein? 😩 UniProt and Google AI have teamed up to create a natural language processing model that has generated over 40 million protein annotations to address this challenge. ebi.ac.uk/about/news/tec…

Predicting the effects of missense variants is a central problem in human genome interpretation. We are thrilled to share our preprint on using cross-protein transfer (CPT) learning to improve zero-shot prediction of disease variant effects: doi.org/10.1101/2022.1… (1/8)