Critical Assessment\nof Protein Engineering (CAPE):\nA Student Challenge on the Cloud

The success of AlphaFold in protein structure prediction\nhighlights\nthe power of data-driven approaches in scientific research. However,\ndeveloping machine learning models to design and engineer proteins\nwith desirable functions is hampered by limited access to high-quality\ndata sets and experimental feedback. The Critical Assessment of Protein\nEngineering (CAPE) challenge addresses these issues through a student-focused\ncompetition, utilizing cloud computing and biofoundries to lower barriers\nto entry. CAPE serves as an open platform for community learning,\nwhere mutant data sets and design algorithms from past contestants\nhelp improve overall performance in subsequent rounds. Through two\ncompetition rounds, student participants collectively designed >1500\nnew mutant sequences, with the best-performing variants exhibiting\ncatalytic activity up to 5-fold higher than the wild-type parent.\nWe envision CAPE as a collaborative platform to engage young researchers\nand promote computational protein engineering.