A simplified homology-model builder toward highly protein-like structures: An inspection of restraining potentials

Tae Rae Kim, Sangho Oh, Joshua Sungwoo Yang, Sanghyuk Lee, Seokmin Shin, Jinhyuk Lee

Research output: Contribution to journalArticlepeer-review

43 Scopus citations

Abstract

A homology model builder using simple restraining potentials based on spline-interpolated quadratic functions is developed and interfaced with CHARMM package. The continuity and stability of the potential function were validated, and the parameters were optimized using the CASP7 targets. The performance of the model builder was benchmarked to the Modeller program using the template-based modeling targets in CASP9. The benchmark results show that, while our builder yields the structures with slightly lower packing, backbone, and template modeling scores, our models show much better protein-like scores in terms of normalized discrete optimized protein energy, dipolar distance-scaled finite-ideal gas reference, Molprobity clash, Ramachandran appearance Z-score, and rotamer Z-score. As our model builder is interfaced with CHARMM, it is advantageous to directly use other CHARMM functionality and energy functions to refine the model structures or to use the models for other computational studies using CHARMM.

Original languageEnglish
Pages (from-to)1927-1935
Number of pages9
JournalJournal of Computational Chemistry
Volume33
Issue number24
DOIs
StatePublished - 15 Sep 2012

Keywords

  • CASP
  • CHARMM
  • homology modeling
  • protein-like scores
  • restraining potential

Fingerprint

Dive into the research topics of 'A simplified homology-model builder toward highly protein-like structures: An inspection of restraining potentials'. Together they form a unique fingerprint.

Cite this