AlphaFold - ML for protein structure prediction

AlphaFold was a revolution for structural biology. Here, Deniz Akpinaroglu discusses the principles that make it possible to go from protein sequence to 3D structural coordinates. This includes: data representations (sequence, evolutionary information, secondary structure, protein surface, atomic coordinates, protein graph), tokens, bias reduction techniques, restraints in folding simulations, and model confidence. The AlphaFold architecture is also specifically analyzed to identify the elements that impact its game-changing accuracy.
Protein Modeling and Design with PyRosetta and Machine Learning
• Protein Modeling and D...
AlphaFold2 paper: www.nature.com/articles/s4158...
Video from the Rosetta Commons RaMP Bootcamp (July 2023)
Instructor: Deniz Akpinaroglu (UCSF)
Credits:
Instructor: Deniz Akpinaroglu (UCSF)
RaMP Director and Rosetta Commons Director: Jeffrey Gray (JHU)
RaMP Program Administrator:Camille Mathis (JHU)
Rosetta Commons Instructional Designer: Ashley Vater (UC Davis)
Video Production: Elizabeth Bonilla (JHU)
Funding: Rosetta Commons, National Science Foundation, and Johns Hopkins University