Table 2.
Method | CASP14 group name | CASP14 resultsa | Description; URL address |
---|---|---|---|
D-I-TASSER | Zhang-Server | First Place Server | Template and deep learning distance/orientation/hydrogen bond network-guided folding; https://zhanglab.dcmb.med.umich.edu/D-I-TASSER/ |
D-QUARK | QUARK | Second Place Server | Deep learning distance/orientation-guided folding; https://zhanglab.dcmb.med.umich.edu/D-QUARK/ |
AlphaFold2 | AlphaFold2 | First Place Human Group | End-to-end deep learning-based model prediction; https://github.com/deepmind/ |
Rosetta | BAKER | Second Place Human Group | Deep learning distance/orientation-guided folding; Robetta Server: https://robetta.bakerlab.org trRosetta Server: https://yanglab.nankai.edu.cn/trRosetta/ |
Methods in CASP are divided into server and human groups. Predictions by server groups are fully automated, whereas those by human groups do not have to be.