For the current REF see the REF 2021 website REF 2021 logo

Output details

8 - Chemistry

University of York

Return to search Previous output Next output
Output 40 of 191 in the submission
Name of software

COOT : Crystallographic Object-Oriented Toolkit

Type
G - Software
Name of software house
MRC Laboratory of Molecular Biology, Cambridge
Year
2011
Number of additional authors
0
Additional information

Since 2008, the following research has enabled new functionality in the Coot software:

1. We have developed a new scoring function enabling rapid comparison of protein fragments against a database of known structures. We linked this to a purpose-built fragment database of 200,000 amino acids. The resulting tools allow users to build missing loops into protein structures in real time, build common motifs into a model and check model fragments against existing structures. The development was evaluated by user feedback.

2. When data resolution is poor, electron density often provides too little information to allow model refinement with conventional bond length and angle restraints. New restraints are required to maintain realistic geometry in such cases. We implemented a Ramachandran restraint based on a Bayesian statistical score to encourage physically realistic conformations. We also implemented a set of empirically determined, main-chain restraints to preserve secondary structure features. These were evaluated both by user feedback, and on their ability to restore known structure features of test structures in the presence of minimal electron density.

3. The very first assessment of the quality of an electron density map looks for common motifs such as α-helices and β-strands. We therefore examined an ensemble of structures to identify key 'fingerprints' that may be used to detect such motifs. We implemented a search algorithm to place secondary structure features on the basis of the fingerprints. The tool now provides a near-instantaneous visual evaluation of the electron-density map.

4. Interpretation of noisy electron-density maps is facilitated by inspection of the electron density at different levels of detail, requiring re-calculation of the map. We developed a cacheing mechanism enabling retention of the required information in the program allowing the user to vary map sharpening in real time. The feature has received very positive user feedback.

Interdisciplinary
-
Cross-referral requested
-
Research group
B - Chemical and Structural Biology (YSBL)
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-