https://predictioncenter.org/index.cgi
CASP 16 is underway
Please describe CASP and how Alphafold fits in
[AI]: CASP (Critical Assessment of protein Structure Prediction) is a biennial scientific experiment and a community-wide effort aimed at gauging the progress of computational methods in predicting protein three-dimensional structures. Initiated in 1994, CASP brings together computational biologists and experimentalists in a collaborative competition format to assess the accuracy of protein structure prediction methods. Participants are provided with the amino acid sequences of proteins whose structures have not yet been publicly disclosed, and their task is to predict the three-dimensional structures of these proteins. The predictions are then compared to the experimentally determined structures, which are revealed at the end of the competition, providing a stringent benchmark for evaluating the prediction methodologies.
AlphaFold, developed by DeepMind (a subsidiary of Alphabet Inc., Google's parent company), represents a significant advancement in the field of computational biology, particularly in the context of CASF competitions. AlphaFold first made a resounding impact at CASP13 (held in 2018), demonstrating remarkable accuracy in protein structure prediction. Its success was even more pronounced at CASP14 (held in 2020), where AlphaFold achieved unprecedented levels of accuracy, far surpassing other methods. The system uses a deep learning approach, incorporating a novel architecture that leverages spatial graph representations of protein structures. By predicting the distances and angles between the amino acids in a protein, AlphaFold can accurately model the complex three-dimensional structure of proteins.
AlphaFold's achievements in CASP have been hailed as a milestone in computational biology, likened to solving a 50-year-old grand challenge in biology. Its success has not only validated the potential of artificial intelligence (AI) in scientific discovery but also opened new avenues for protein research. This breakthrough holds immense implications for understanding biological processes, developing new therapies, and designing synthetic proteins for industrial, environmental, and pharmaceutical applications.
In summary, within the framework of CASP, AlphaFold fits as a high-water mark in the progress of protein structure prediction, symbolizing a shift towards the increasingly prominent role of AI and machine learning in tackling complex biological problems.