Baylor College of Medicine Wah Chiu*, PI Grigore Pintilie* Matthew Baker* Matthew Dougherty Steven Ludtke Rutgers University Helen Berman, co-PI Catherine Lawson* Raul Sala* Brian Hudson John Westbrook European Bioinformatics Institute Gerard Kleywegt, co-PI Eduardo Sanz Garcia* Matthew Conroy* Ingvar Lagerstedt Ardan Patwardhan EMDataBank Project Team *supported by NIH R01GM079429
9
Embed
Baylor College of Medicine Wah Chiu*, PI Grigore Pintilie* Matthew Baker* Matthew Dougherty Steven Ludtke Rutgers University Helen Berman, co-PI Catherine.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
• Create single archival repository for maps, models and metadata
• Provide web-based access tools to deliver and manage maps, models and metadata
• Assess and develop tools needed for map and model validation
• Charge: advise on approaches to validate EM maps and map-derived models
• Sept 28-29, 2010 meeting
• Day 1: plenary presentations
• Day 2: separate Map and Model group meetings to discuss specific issues
EM Validation Task Force
Co-chairs:Richard HendersonAndrej Sali
EM VTF MembersMaps Richard Henderson-Chair (MRC) Bridget Carragher (Scripps) Ken Downing (LBL) Edward Egelman (U Virginia) Joachim Frank (Columbia) Niko Grigorieff (Brandeis) Wen Jiang (Purdue) Steven Ludtke (Baylor) Ohad Medalia (Ben Gurion) Pawel A. Penczek (UT Houston) Michael Schmid (Baylor) Alasdair Steven (NIH)
Models Andrej Sali-Chair (UCSF) Matthew Baker (Baylor) Peter Rosenthal (NIMR) Michael G. Rossmann (Purdue) Gunnar Schroeder (FZ Juelich) Willy Wriggers (DE Shaw)
General Recommendations Journals should encourage map and/or model deposition
before publication• A set of guidelines for what should be included with the
publication of a 3DEM structure should be established and shared with journal editors
• Maps and models that have been peer reviewed and referred to in published literature should be publicly available
EMDataBank can play a key role• Provide unified access to 3DEM molecular models and their
annotations• Support development of data standards to facilitate exchange
of information and algorithms• User interface should allow a broad range of queries
Map Recommendations - 1 Standards for assessing map resolution and accuracy
need to be developed• No widely accepted validation methods to assess map
accuracy• EMDataBank can provide table of existing map-validation
techniques Resolution should be reported and visible structural
features should be in accordance with the claimed resolution
• Full FSC curve to Nyquist frequency, map manipulations and transformations as well as key parameters of reconstruction algorithms should be reported
• Map symmetry should be validated
Map Recommendations - 2
Depositions should include annotations specific to each map type
• 2D crystal*, helical*, single particle, tomogram, sub-tomogram average (*may use single-particle methodology)
The requirements and practicalities for archiving raw image data should be investigated
Benchmark raw image data should be made available for methods development
Model Recommendations I
Criteria for assessing models must be established • Methods are being developed, no dominant ones have emerged• Three assessment types:
without regard to the 3DEM map: use X-ray VTF criteria with regard to the 3DEM map: composite scores,
correlation coefficients, real-space fit values with regard to additional data not used in modelling, e.g.