본문 바로가기

DeepFold: enhancing protein structure prediction through optimized loss functions, improved template features, and re-optimized energy function

문서 유형
연구출판물 · 학술지 논문
등재
SCIEScopus
저널 순위
JCR Q1Scopus Q1
저널
BIOINFORMATICS39(12) · 2023
피인용
Web of Science 21Scopus 21
FWCI (Scopus)
1.81
JCR IF
JCR 2022 5.8
ISSN
1367-4803

저자

  1. Lee, Jae-Won제1저자Hanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  2. Won, Jong-HyunHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  3. Jeon, SeonggwangHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  4. Choo, YujinKorea Inst Adv Study, Ctr Adv Computat, Seoul 02455, South Korea
  5. Yeon, YubinHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  6. Oh, Jin-SeonKorea Inst Adv Study, Ctr Adv Computat, Seoul 02455, South Korea
  7. Kim, MinsooKyungil Univ, Sch Elect Engn, Gyongsan 02841, South Korea
  8. Kim, SeonhwaStandigm Inc, Seoul 06234, South Korea
  9. Joung, InsukHanyang Univ, Artificial Intelligence Inst, Seoul 04763, South Korea
  10. Jang, CheongjaeChangwon Natl Univ, Basic Sci Res Inst, Chang Won 51140, South Korea
  11. Lee, Sung JongPusan Natl Univ, Sch Comp Sci & Engn, Pusan 46241, South Korea
  12. Kim, Tae HyunHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  13. Jin, Kyong HwanStandigm Inc, Seoul 06234, South Korea
  14. 송길태Korea Inst Adv Study, Sch Computat Sci, Seoul 02455, South Korea
  15. Kim, Eun-SolHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  16. Yoo, Jejoong
  17. Paek, EunokHanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  18. Noh, Yung-Kyun교신저자Hanyang Univ, Dept Comp Sci, Seoul 04763, South Korea
  19. Joo, Keehyoung교신저자Korea Inst Adv Study, Ctr Adv Computat, Seoul 02455, South Korea

초록

MOTIVATION: Predicting protein structures with high accuracy is a critical challenge for the broad community of life sciences and industry. Despite progress made by deep neural networks like AlphaFold2, there is a need for further improvements in the quality of detailed structures, such as side-chains, along with protein backbone structures. RESULTS: Building upon the successes of AlphaFold2, the modifications we made include changing the losses of side-chain torsion angles and frame aligned point error, adding loss functions for side chain confidence and secondary structure prediction, and replacing template feature generation with a new alignment method based on conditional random fields. We also performed re-optimization by conformational space annealing using a molecular mechanics energy function which integrates the potential energies obtained from distogram and side-chain prediction. In the CASP15 blind test for single protein and domain modeling (109 domains), DeepFold ranked fourth among 132 groups with improvements in the details of the structure in terms of backbone, side-chain, and Molprobity. In terms of protein backbone accuracy, DeepFold achieved a median GDT-TS score of 88.64 compared with 85.88 of AlphaFold2. For TBM-easy/hard targets, DeepFold ranked at the top based on Z-scores for GDT-TS. This shows its practical value to the structural biology community, which demands highly accurate structures. In addition, a thorough analysis of 55 domains from 39 targets with publicly available structures indicates that DeepFold shows superior side-chain accuracy and Molprobity scores among the top-performing groups. AVAILABILITY AND IMPLEMENTATION: DeepFold tools are open-source software available at https://github.com/newtonjoo/deepfold.

키워드

CONTACT PREDICTIONCOEVOLUTIONREFINEMENTSIMULATION

식별자

Scopus EID
2-s2.0-85179019341
WoS UID
001117573600001