Single Particle Reconstruction with EMAN GroEL Methods for High Resolution Refinement Donghua Chen in Single Particle Processing Joanita Jakana Wah Chiu Steve Ludtke Jiu-Li Song (UT-SW Med) David Chuang (UT-SW Med) Asst. Professor, Biochemistry, BCM Co-director,NCMI NCRR EMAN: http://ncmi.bcm.tmc.edu/eman GroEL 2000 (15 Å) GroEL 2001 (11.5 Å) 5000 particles, JEOL 4000 5000 particles, JEOL 4000 GroEL 2003 (6 Å) 2005 30,000 particles, JEOL 2010F
Animation Unavailable in PDF version Jeol 3000 7 Days of imaging, 910 micrographs 1.06 Å/pix, Nikon 9000 scanner 135 used, 34,868 particles Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version
Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version
Animation Unavailable in PDF version Animation Unavailable in PDF version Animation Unavailable in PDF version Ca 2+ Release Channel Irina Serysheva Wah Chiu Susan Hamilton 200 kV image of ice-embedded RyR1 (no continuous CF) Ca 2+ Release Channel Myofibril Plasma membrane T-tubule • SR membrane, triggered by DHPR in T-tubule Terminal cisternae of SR • Homotetramer • ~2200 kDa • Releases Ca ++ which initiates Tubules of sarcoplasmic cross-bridge cycle reticulum 1.2 µ m 500 Å
~30 Å Resolution Animation Unavailable in PDF version 20 Å Resolution 14 Å Resolution 9.6 Å Resolution Animation Unavailable in PDF version
Animation Unavailable in PDF version 270 Å T-tubule membrane cytoplasmic face Cytoplasm 190 Å } TM SR membrane SR lumen SR lumenal face Sequence assignment of observed helices Sequence assignment of observed helices Filter Hinge Filter Hinge RyR1: RyR1: 4864 –NKSEDEDEPDMKCDDMMTCYLFHMYVGVRAGGGIGDEIEDPAGDEYELYRVVFDITFFFFVIVILLAIIQGLIIDAFGELRDQQEQVKEDMETK- 4957 4864 –NKSEDEDEPDMKCDDMMTCYLFHMYVGVRAGGGIGDEIEDPAGDEYELYRVVFDITFFFFVIVILLAIIQGLIIDAFGELRDQQEQVKEDMETK- 4957 M9 M10 M9 M10 Helix 2 Helix 1 Filter KcsA: 36 -QLITYPRALWWSVETATTVGYGDLYPVTLWGRCVAVVVMVAGITSFGLVTAALATWFVGREQ -119 Pore helix Inner helix lumenal side (‘out’) Filter Hinge MthK: 45 -SWTVSLYWTFVTIATVGYGDYSPSTPLGMYFTVTLIVLGIGTFAVAVERLLEFLINREQ- 103 Pore helix Inner helix Pore helix Filter kink Filter Pore helix Helix 2 Helix 1 kin k Inner helix Inner helix MthK RyR1 RyR1/KcsA/MthK KcsA CCD vs. Film CCD Film + Scanner Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Align and Particle Classify Average Images Particles Classes
Initial 3D Uniform Build New Final 3D Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Model Projections 3-D Model Model Align and Align and Particle Classify Particle Classify Average Average Images Particles Images Particles Classes Classes Initial 3D Final 3D Initial 3D Final 3D Uniform Build New Uniform Build New Model Projections 3-D Model Model Model Projections 3-D Model Model Align and Align and Particle Classify Particle Classify Average Average Images Particles Images Particles Classes Classes Refine from Gaussian Ellipsoid Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Align and Particle Classify Average Images Particles Classes
Iteration 1 Iteration 2 Iteration 3 Iteration 4 Iteration 5 How do we get to Higher Resolutions? • Get a better microscope • Find a better microscopist • Algorithm Improvements
M(s) = F(s) C(s) 2 E(s) 2 + N(s) 2 C(s) E(s) + N(s) N(s) M(s) 2 F(s) C(s) 2 E(s) 2 + N(s) 2
Image Classification � ? � � ? � � ? � Alignment/Registration Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Align and Particle Classify Average Images Particles Classes
Alignment/Registration Alignment/Registration Alignment/Registration Initial 3D Final 3D Uniform Build New Model Projections 3-D Model Model Align and Particle Classify Average Images Particles Classes Measures of Similarity • Correlation Coefficient • Variance (transformed density) ( - ) 2 = • Variance (matched filter) • Phase Residual • Mutual Information • etc.
( - ) 2 = ( - ) 2 = And the Answer is… • Wiener filter particle • Filter reference to match ( - ) 2 = • Normalize reference density to particle • Calculate variance Model Bias Base Noisy Align to Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Align and Particle Classify Average Images Particles Classes 25 100 250 1000 2000
Model Bias Model Bias Base Base Noisy (~10% contrast) Align to Noisy (~10% contrast) Align to 25 100 250 1000 2000 25 100 250 1000 2000 Model Bias Model Bias Base Noisy Align to Base Noisy Align to Iter x4 25 100 250 1000 2000 25 100 250 1000 2000 Model Bias Model Bias Base Noisy Base Noisy (~10% contrast) Align to Align to Iter x8 25 100 250 1000 2000 25 100 250 1000 2000
Model Bias Base Noisy Align to Initial 3D Uniform Build New Final 3D Model Projections 3-D Model Model Iter x4 Align and Particle Multi-Classify Average Images Particles Classes 25 100 250 1000 2000 The Future • Each particle -> best n classes • Better similarity criteria • More restrictive exclusion from class-avg • Improved CTF model • Per-particle CTF (at least defocus) • Beam tilt • Better 3-D reconstruction • New refinement methodologies
Recommend
More recommend
