1 / 38

Molecular replacement and model-building using distant homology models as templates

Molecular replacement and model-building using distant homology models as templates. ACA Trueblood Award Lecture July 24, 2013 Honolulu, Hawaii. Tom Terwilliger Los Alamos National Laboratory. Kenneth Trueblood 1920-1998.

aran
Download Presentation

Molecular replacement and model-building using distant homology models as templates

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Molecular replacement and model-building using distant homology models as templates ACA Trueblood Award Lecture July 24, 2013 Honolulu, Hawaii Tom Terwilliger Los Alamos National Laboratory Kenneth Trueblood 1920-1998

  2. The challenge: Crystal structure determination from a distant homology model The problems: Search model too different from target to find its location by molecular replacement Correctly-placed model too different from target to yield useful electron density maps

  3. The challenge: Crystal structure determination from a distant homology model Solutions using Rosetta: Homology modeling before molecular replacement Modeling with density at early stages of structure rebuilding DiMaio et al. (2011). Improving molecular replacement by density and energy guided protein structure optimization. Nature 473, 540-543

  4. Complementarity of PHENIX and Rosetta model-building Characteristic Optimization Model-building approach Fragment libraries Model-building target Refinement target Crystallographic model-building (PHENIX) Interpretation of patterns of density Density search for regular secondary structure 3-residue fragment library Fit to density Structure-factor likelihood refinement target Structure-modeling (Rosetta) Creating physically plausible models Ab-initio modeling or homology modeling 3- and 9-residue libraries Rosetta force field (optional density term) Rosetta force field (optional density term)

  5. Generating improved homology models for molecular replacement with Rosetta ag9603 NMR model, 100% identity 1.7 Å data

  6. NMR model of of ag9603 NMR model (CC to final map: 0.17) Final model

  7. Typical Rosetta model of of ag9603 Rosetta model 791 (CC to final map: 0.27) Final model

  8. Best Rosetta model of of ag9603 Best Rosetta model for ag9603 (CC to model map: 0.34) Final model

  9. Improving models with Rosetta using density Density fit as part of optimization targetModeling segments not in templateComparison with simulated annealing

  10. Rosetta models vs SA models

  11. Integrating Rosetta modeling in Phenixphenix.mr_rosettaMolecular replacementRebuilding with RosettaRebuilding with phenix.autobuild

  12. mr_rosetta example: HP3342, 22% identity template, 3.2 Å data, B=87 A2

  13. hp3342, 22% identity template (1vgy), 3.2 Å dataDensity- modified map based on 1vgyYellow: final modelBlue: template (1vgy)

  14. MR_rosetta example: hp3342

  15. hp3342, 22% identity template (1vgy), 3.2 Å dataDensity- modified map based on 1vgyYellow: final modelBlue: template (1vgy) Pink: Highest-scoring Rosetta model

  16. hp3342, 22% identity template (1vgy), 3.2 Å dataDensity- modified map based on Rosetta modelYellow: final modelBlue: template (1vgy) Pink: Highest-scoring Rosetta model

  17. hp3342, 22% identity template (1vgy), 3.2 Å dataDensity- modified map based on Rosetta modelYellow: final modelGreen: autobuild model

  18. HP3342, 22% identity template, 3.2 Å data, R/Rfree=0.34/0.41 B=87 A2

  19. mr_rosetta example II: XMRV, 30% identity template (2hs1), 2.0 Å data

  20. XMRV, 30% identity template, 2.0 Å dataDensity- modified map based on 2hs1Yellow: final modelBlue: template (2hs1) Pink: Highest-scoring Rosetta model

  21. MR_rosetta example: XMRV

  22. XMRV, 30% identity template, 2.0 Å dataDensity- modified map based on 2hs1Yellow: final modelBlue: template (2hs1) Red: Highest-scoring Rosetta model

  23. XMRV, 30% identity template, 2.0 Å dataDensity- modified map from Rosetta modelYellow: final modelBlue: template (2hs1) Red: Highest-scoring Rosetta model

  24. XMRV, 30% identity template, 2.0 Å dataDensity- modified map from Rosetta modelYellow: final modelBlue: template (2hs1) Purple: Relaxed Rosetta model

  25. XMRV, 30% identity template, 2.0 Å dataDensity- modified map from autobuildYellow: final modelBlue: template (2hs1) Green: autobuild model

  26. XMRV, 30% identity template, 2.0 Å dataDensity- modified map from autobuildYellow: final modelBlue: template (2hs1) Green: autobuild model

  27. Structure determination of cab55348 (using template supplied by user) 1.9 A, 28% sequence identity(AutoMR alone fails with R/Rfree=0.47/0.53) MR model: blue, Final model: green

  28. MR model : blue Final model: pink Rosetta models cycle 1: green Sample Rosetta models in cycles 1 and 2,

  29. MR model : blue Final model: pink Map from refined MR model Rebuilding in a poor section of the starting model

  30. MR model : blue Final model: pink Rosetta models cycle 1: green Map from refined MR model Rebuilding in a poor section of the starting model

  31. MR model : blue Final model: pink Best-scoring Rosetta model cycle 1: yellow Map from refined MR model Rebuilding in a poor section of the starting model

  32. MR model : blue Final model: pink Rosetta models cycle 2: yellow AutoBuild Map from cycle 1 Rebuilding in a poor section of the starting model

  33. AutoBuild model cycle 2 Map CC: 0.78 R/Rfree = 0.26/0.31 Rebuilding in a poor section of the starting model

  34. mr_rosetta rebuilding starting with placed templates

  35. The challenge: Crystal structure determination from a distant homology model Another solution: Morphing (distortion) of the template using a density map to make it more like the target structure

  36. Related structures often have high local similarity ag9603; approximate NMR model as template in pink

  37. Related structures often have high local similarity XMRV PR, 30% identity template (2hs1) in blue

  38. Related structures often have high local similarity cab55348 32% identical template (Cip2) in blue

More Related