Nature Communications (May 2021)

CopulaNet: Learning residue co-evolution directly from multiple sequence alignment for protein structure prediction

  • Fusong Ju,
  • Jianwei Zhu,
  • Bin Shao,
  • Lupeng Kong,
  • Tie-Yan Liu,
  • Wei-Mou Zheng,
  • Dongbo Bu

DOI
https://doi.org/10.1038/s41467-021-22869-8
Journal volume & issue
Vol. 12, no. 1
pp. 1 – 9

Abstract

Read online

Protein structure prediction is a challenge. A new deep learning framework, CopulaNet, is a major step forward toward end-to-end prediction of inter-residue distances and protein tertiary structures with improved accuracy and efficiency.