AI-powered protein design

Deep learning for protein design is one of the seven technologies to watch in 2024, according to Nature Journal, and the recent Nature Biotechnology also put the spotlight on this topic.

What is protein design? How can we use AI to optimize existing proteins and design completely new ones? What AI-based methods have been developed in this field? What are the challenges? Many articles have answered these questions very nicely. In this newsletter, we list these articles and resources (especially those published in recent weeks) for readers to explore further.

• What is protein design? A beginner’s guide to protein design written by the most famous academic institute in this field, the Institute for Protein Design at the University of Washington.

• List of papers about protein design using deep learning. A GitHub repository hosting the most up-to-date list of papers in this field. This is a great resource to quickly navigate the milestone papers on different topics about protein design.

• De novo protein design—From new structures to programmable functions. A review recently published in Cell highlights the emerging approaches and challenges for de novo protein design.

• Generative AI for Controllable Protein Sequence Design. A survey of generative AI models and optimization algorithms for protein design.

• A Survey of Generative AI for De Novo Drug Design: New Frontiers in Molecule and Protein Generation with corresponding GiHub repository, GenAI4Drug.

• Protein design articles within Nature. A collection of papers published in Nature journals.

• Focus on protein engineering. The recent issue of Nature Biotechnology, which published many reviews on this topic, e.g.,

  • Protein engineering and design. Recent patents relating to systems and methods for designing engineered polypeptides.

  • Sparks of function by de novo protein design. A review that examines the advances of AI-based protein design in the broader context of classical de novo protein design.

  • Machine learning for functional protein design. A review focusing on the machine learning methods and classifying them on the basis of their use of three core data modalities: sequences, structures, and functional labels.

  • What does it take for an ‘AlphaFold Moment’ in functional protein engineering and design?

If you find the newsletter helpful, please consider:

• 🔊 Sharing the newsletter with other people

• 👍 Upvoting on Product Hunt

• 📧 Sending any feedback, suggestions, and questions by directly replying to this email or writing reviews on Product Hunt

• 🙏 Supporting us with a cup of coffee.

Thanks, and see you next time!