Outcome of Crash Course Training on Protein Structure Prediction with Artificial Intelligence
Balamurugan, D.; Dougherty, M.; Lubin, J.; Arias, P.; Chang, J.; Dalenberg, K.; Kholodovych, V.; Zelzion, E.; Khare, S.; von Oehsen, J. B.; Zwick, M. E.; Burley, S. K.
Show abstract
Protein structure predictions have broad impact on several science disciplines such as biology, bioengineering, and medical science. AlphaFold2[1] and RoseTTAFold[2] are the current state-of-the-art AI methods to predict the structures of proteins with an accuracy comparable to lower-resolution experimental methods. In its 2021 year review, both these methods were recognized as "breakthrough of the year" by Science magazine[3] and "method of the year" by Nature magazine [4]. It is timely and important to provide training and support on these emerging methods. Our crash course "Enabling Protein Structure Prediction with Artificial Intelligence "was conducted in collaboration with domain experts and research computing professionals. The crash course was well received by the community as there were 750 registrants from all over the world. Here we provide the summary of the crash course, describe our findings in organizing the crash course, and explain what preparation steps helped us with the hands-on training. CCS CONCEPTSComputing methodologies a Machine learning a Machine learning approaches a Bio-inspired approaches
Matching journals
The top 6 journals account for 50% of the predicted probability mass.