doi.bio/tamas_berghammer

Tamas Berghammer link

Tamas Berghammer is a researcher in the field of biochemistry and genomics. He is known for his work on the AlphaFold project, which focuses on developing an accurate protein structure prediction algorithm.

Education and Career

Berghammer is affiliated with DeepMind Technologies Limited and has filed patents related to machine learning for predicting protein structures. He collaborated with other researchers at DeepMind, Google, and Alphabet to develop the AlphaFold algorithm.

Notable Works

• Highly accurate protein structure prediction with AlphaFold link(https://www.nature.com/articles/s41586-021-03819-2) (2021): This paper introduces the AlphaFold model, which predicts protein structures with high accuracy. The method incorporates novel neural network architectures and training procedures based on evolutionary, physical, and geometric constraints. The work was published in Nature and validated through the Critical Assessment of Protein Structure Prediction (CASP14).
• Highly accurate protein structure prediction for the human proteome link(https://www.nature.com/articles/s41586-021-03828-1) (2021): This paper applies the AlphaFold model to predict structures for the entire human proteome. It demonstrates the scalability and potential impact of the algorithm in structural bioinformatics.
• Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13) link(https://www.sciencedirect.com/science/article/pii/S0887358519300083) (2019): This work describes the use of multiple deep neural networks to predict protein structures in the CASP13 competition. The team, including Berghammer, achieved notable results, laying the foundation for further improvements in CASP14.

Co-Authors

John Jumper, Richard Evans, Alexander Pritzel, Demis Hassabis, and others.