Related Experiment Videos
Mismodeled purines: implicit alternates and hidden Hoogsteens
Bradley J Hintze1, Jane S Richardson1, David C Richardson1
1Department of Biochemistry, Duke University, Durham, NC 27710, USA.
Acta Crystallographica. Section D, Structural Biology
|October 11, 2017
Summary
This study investigated rare Hoogsteen base pairs in DNA crystal structures, finding modeling errors are infrequent. The occurrence of Hoogsteen base pairs in DNA remains around 0.3%.
Area of Science:
- Structural Biology
- Biochemistry
- Crystallography
Background:
- Hoogsteen base pairs are rarely observed in DNA crystal structures.
- Rare conformations in DNA can be subject to under- or over-modeling issues.
- Identifying syn/anti purine flips is crucial for accurate structural analysis.
Purpose of the Study:
- To investigate potential under- or over-modeling of Hoogsteen and other syn purines in DNA crystal structures.
- To identify and validate instances of purine base pair flips in DNA crystallography.
- To assess the frequency and impact of modeling errors on DNA structure determination.
Main Methods:
- Difference electron-density peaks were used to identify candidate purines requiring syn/anti flips.
- Manual inspection and rebuilding/refinement of 20 candidate cases were performed.
- Analysis of DNA sequences, protein binding, and crystallographic data.
Main Results:
- 14 out of 20 candidate purine flips were confirmed as authentic.
- Seven Watson-Crick base pairs were incorrectly modeled and should be Hoogsteen, and vice versa for three cases.
- Five cases involved unmodeled alternate duplexes in semi-palindromic DNA sequences bound by homodimeric proteins.
Conclusions:
- Modeling errors for Hoogsteen base pairs and syn/anti flips are infrequent in DNA crystal structures.
- The bias towards modeling anti purines is minimal.
- The overall occurrence rate of Hoogsteen base pairs in DNA crystal structures is approximately 0.3%.