Using network community clustering algorithms [...]
summary
Claim: automatic community clustering of secondary-structure elements produces compact rigid-body groups that improve molecular replacement and reduce model error in practice. Implementation uses Python igraph to cluster a graph whose nodes are helices/strands and edges encode distances, then feeds groups into archin Balo / archin BOS reader fragment refinement and rotation/translation searches, yielding examples where models move ~0.47 angstrom closer to solved structures and rescue difficult low–sequence-identity cases.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:26Proteins and structure
- 7:47Experimental methods
- 13:30Phase problem in crystallography
- 17:01Non-standard molecular replacement
- 20:14Protein structures as graphs
- 23:01Community clustering algorithms
- 26:59Results and applications
- 28:45Future improvements
- 29:48Acknowledgments
- 30:29Q&A
related talks
21mHierarchical Clustering: Finding the awkward reunions
31mSimple Models and Exact Algorithms for Computing Network Modules
01h 06mIntroduction to Biological Network Analysis III: Identifying Network Modules
49mIntroduction to Biological Network Analysis IV: Network Alignment and Querying
47mPhase Transitions in Semidefinite Relaxations (A Fast & Robust Algorithm for Community Detection)
58mIntroduction to Biological Network Analysis I: Network Basics and Properties
35mClustering: A Guide for the Perplexed
47mCrowdsourcing for Clustering: Algorithms and Information-Theoretic Lower Bounds