Spectrometer Format Conversion

1
Spectrometer Format Conversion 1D-4D Fourier
Transform and Signal Enhancement Spectral
Visualization 1D-4D Peak Detection and
Quantification Position, Amplitude, Width and
Modulation/Evolution Spectral Assignment Extractio
n of Structural Parameters Molecular Structure
Calculation Molecular Display and Structure
Verification Exploitation of Structure Screening,
Automation, Process Control Spectral Imaging
Ad Bax ? Joeseph Barchi ? James Chou ? Gabriel
Cornilescu ? Alex Grishaev Stephan Grzesiek
? Georg Kontaxis ? John Kuszewski ? Ryan McKay ?
John Pfeiffer Michael Shapiro ? Tobias Ulmer ?
Gerteen Vuister ? Justin Wu ? Shen Yang
Guang Zhu ? Edward Zartler
EMBO 2009 Biomolecular NMR
Frank Delaglio Course Site www.nmrscience.com/emb
o
2
NMR Signal Processing
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
Discrete Fourier Transform
16
The Fourier Transform
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
/ Fourier transform of complex data tR,tI to
produce fR,fI. / void ft( float tR, float tI,
float fR, float fI, int size ) float vR,
vI, twoPI int mid, k, n twoPI
4.0acos( 0.0 ) mid size/2 for( k
0 k lt size k ) / For every
output freq point / fRk
0.0 fIk 0.0 for( n 0 n
lt size n ) / Sum over input times
sinusoid. / f
twoPI(k - mid)n/size vR cos( f
) vI sin( f ) fRk
tRnvR - tInvI fIk
tRnvI tInvR
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
Repeat 1D experiment for increasing values of
t1 Measure cos and sin modulated versions at each
value of t1
26
2D HN/N HSQC 1D Signals are Dispersed in Two
Dimensions
27
(No Transcript)
28
(No Transcript)
29
NMRPipe
30
(No Transcript)
31
NMRPipe Spectral Processing as a UNIX Pipeline

• NMR Parameter Calculation (Shifts, Dipolar
  Couplings, PCS, etc)
• Customization is via standard scripting languages
  (C-shell, TCL)
• Created and Maintained by one developer, with
  contributed modules
• SunOS, Solaris, IRIX, HP/UX, DEC OSF, IBM AIX,
  Convex OS, Cray OS, Mac OS X, Linux, WindowsXP
  Interix
• Bottom-up Software Design

• 1D-4D FT, LP, MEM, FDM, Parallel Processing
• 1D-4D Peak Detection and Quantification
• Spectral Graphics, Extracts, Strips, Projections
• Extensively Customizable (Custom Processing
  Functions, Automated and Interactive Schemes)
• Molecular Structure Calculation

32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
NMRWish

• Customized version of TCL/TK wish interpreter
• Script-based Interactive Spectral graphics
  (multi-window and PostScript)
• Generic Database Engine (GDB)
• Manipulate Peak Data, Assignments, NMR
  Parameters, and Molecular Structure

36
(No Transcript)
37
(No Transcript)
38
Special Methods in NMR Signal Processing
39
Linear Prediction
40
Linear Prediction
FT
FT with LP
41
(No Transcript)
42
(No Transcript)
43
Maximum Entropy Method
44
(No Transcript)
45
Spectral Matrix Decomposition
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
Non-Uniform Sampling
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
Matrix Decomposition and Non-Uniform Sampling
X
X
62
MEM and NUS
63
(No Transcript)
64
(No Transcript)
65
(No Transcript)
66
(No Transcript)
67
Useful Graphics Strategies
68
Graphics Strategies of Edward Tufte www.edwardtuft
e.com
Above all, Show the Data Show Cause and Effect
Represent Data and Scale Faithfully Maximize
Data Ink and Data Density, Minimize Chart
Junk Shrink Graphics - Integrate Text, Values,
and Graphics - Be Multivariate Use Layers Use
Macro and Micro Interpretations - Clarify by
Adding Detail Conserve Color Space Use Small
Multiples Find Ways to Show All of the
Data Treat Design as a Solved Problem, then Find
the Best Examples
69
Debate on the Challenger Launch
Morton Thiokol engineers debated the problem of
O-ring failure due to low temperature for several
hours the night before the launch, and made the
companys only no-launch request in 12 years.
Their presentation of evidence did not convince
NASA management. The shuttle blew up 73 seconds
after ignition.
70
(No Transcript)
71
(No Transcript)
72
(No Transcript)
73
(No Transcript)
74
(No Transcript)
75
Napoleans Route 422,000 Men to 10,000 Men, Five
Dimensions
76
(No Transcript)
77
Bax Group Figure 18 values
78
Weather Statistics 1,800 Values, Four
Variables, Notations
79
.

80
(No Transcript)
81
(No Transcript)
82
(No Transcript)
83
(No Transcript)
84
(No Transcript)
85
(No Transcript)
86
(No Transcript)
87
(No Transcript)
88
(No Transcript)
89
Structural Data from NMR
90
NOE Distance
J-Coupling
Chemical Shift
91
(No Transcript)
92
(No Transcript)
93
2D Homonuclear NMR Correlates Signals in 1D
COSY H-H of adjacent C-C
TOCSY All H-H in Spin System
94
(No Transcript)
95
Identify many H-H short range NOE
distances Supplement with torsions from
J-Coupling values Assume standard peptide
geometry Use simulated annealing to find a
structure which matches distances
96
VARS INDEX RESID_I RESNAME_I ATOMNAME_I RESID_J
RESNAME_J ATOMNAME_J D_LO D_HI FORMAT 4d 5d
6s 6s 5d 6s 6s
9.3f 9.3f 1 3 THR HN 4 GLY
HN 3.380 4.680 2 4 GLY HN
5 GLY HN 3.200 4.500 3 5 GLY
HN 6 LYS HN 3.610 4.910 4
6 LYS HN 7 ILE HN 3.450 4.750 5
6 LYS HN 42 GLU HN 2.670 3.970
6 6 LYS HN 43 GLY HN 1.930
3.230 7 7 ILE HN 8 SER HN
3.990 5.290 8 7 ILE HN 22 CYS
HN 1.970 3.270 9 8 SER HN
9 PHE HN 3.940 5.240 10 8 SER
HN 40 ARG HN 2.090 3.390 11
8 SER HN 42 GLU HN 3.700 5.000 12
9 PHE HN 20 TYR HN 1.990 3.290
13 10 TYR HN 37 ASN HN 3.540
4.840 14 10 TYR HN 38 SER HN
3.050 4.350 15 11 GLU HN 12 ASP
HN 1.820 3.120 etc . . .
97
NOE distances are only qualitative A given peak
might be the only evidence of an interaction A
mis-assigned peak can be similarly problematic
98
(No Transcript)
99
(No Transcript)
100
(No Transcript)
101
2D HN/N HSQC 1D Signals are Dispersed in Two
Dimensions
102
3D 15N and 13C Editing 2D Signals are Dispersed
in Three Dimensions
103
(No Transcript)
104
(No Transcript)
105
(No Transcript)
106
(No Transcript)
107
(No Transcript)
108
(No Transcript)
109
(No Transcript)
110
(No Transcript)
111
(No Transcript)
112
Alternate Approaches to NMR Structure
113
(No Transcript)
114
(No Transcript)
115
(No Transcript)
116
Chemical Shift and Backbone Structure Motif
Match database triplet with target, based on
sum-of squares difference in chemical shifts,
plus residue type homology term. Use central
residue as predictor of phi and psi.
117
(No Transcript)
118
(No Transcript)
119
(No Transcript)
120
Alignment by Liquid Crystal
121
(No Transcript)
122
(No Transcript)
123
Dmax -?0(h/2?) ?i?j/(4?r2r3ij)
124
Dab Dmax (s1)0.5(3.0zIJzIJ - 1.0)
(s2)0.5(xIJxIJ -
yIJyIJ)
(s3)2.0xIJyIJ
(s4)2.0xIJzIJ
(s5)2.0yIJzIJ
125
(No Transcript)
126
After 0.7 hz RMS
Before 7.5 hz RMS
BB RMSD 0.22 A
127
(No Transcript)
128
(No Transcript)
129
(No Transcript)
130
(No Transcript)
131
(No Transcript)
132
Molecular Fragment Replacement (MFR)

• Search PDB for small fragments whose simulated
  dipolar couplings and shifts match the observed
  values.
• Use the fragment information to reconstitute
  larger structural elements.
• Also Sequential NOEs, J values, etc.
• Nucleic Acid Applications

133
(No Transcript)
134
Ideal Backbone Geometry with X-ray Phi/Psi
Values BB RMSD 4.3 A
135
1ubq vs MFR phi/psi refined structure
Initial Structure from Average Phi and Psi of
Fragment Ensemble
136

• MFR Estimation of Tensor Parameters
• Magnitude
• Rhombicity
• Orientation (Euler Angles)

137
(No Transcript)
138
MFR Fragment Tensor Magnitudes Reveal Dynamics
S42
139
Gamma S

• 177 Residues, two similar domains, homologous
  structure is known.
• 179 Amide-Amide NOEs, 70 Methyl-Methyl NOEs,
  including 6 inter-domain
• DC Medium 1 144 HN-N, 111 CA-CB, 150 CA-C, 134
  N-C
• DC Medium 2 147 HN-N, 135 CA-CB, 153 CA-C, 139
  N-C
• Side-chain c1 angles from 3JNCg and 3JCCg
  couplings, c2 from 3JCgCd

140

• Conduct second MFR Search with fixed tensor Da,
  Rh, and relative orientation
• Refine all fragments with fixed tensor Da, Rh
• Phi and Psi for 90 of residues 50 have better
  than 5 degree RMS consenus 33 are 3 degree RMS
  or better.
• Conduct MFR Search with SVD (free tensor)

141
dynReadGMC -gmc gmcDir -pdb pdbName for set i
1 i lt count incr i \
dynSimulateAnnealing -graph -print 50 -rasmol 500
\ -sa stepCount init 100 \
stepCount high 24000 \
stepCount cool 8000 \
timeStep all 3 \
temperature all 4000 \
temperature coolEnd 0 \ -fc dc
coolEnd 2.0 \ torsion
all 50 \ torsion
coolEnd 10 \ noe all
25 \ noe coolEnd
100 \ radGyr all 0.0
set outName format outTemplate i
dynWrite -pdb -src dynInfo(gmc,pdb) -out
outName rem dynInfo(energyText) dynRead
-pdb -src dynInfo(gmc,pdb) -in pdbName
incr iseed 111 srand iseed
142
(No Transcript)
143
MFR Torsions Preserve Secondary Structure During
High Temperature Phase Cooling, MFR Torsion
Restraint Force Constant is Decreased DC Force
Constant is Increased as Ideal Fold is Approached
144
Comparison to 54 identical gB X-ray
Conf. exchange broadened HN
Alignment parameters gel Pf1
Da R Da
R N-domain -11.4 0.38 15.3
0.61 C-domain -11.5 0.42 15.5 0.66
145
SAXS experimental data acquisition
Alex Grishaev
scattering vector
146
SAXS data fit effect on the domain positions
NMR only
NMR SAXS
Xray of gB
147
SAXS data fit effect on structural accuracy
148
SAXS data for Kay/Tugarinov 82 kD MSG
149
SAXS-refined NMR structure of MSG differs by 3.3
vs 4.5 A from X-ray
150
Consistent blind protein structure generation
from NMR chemical shift data Proc Natl Acad Sci
USA, (2008) 105, 4685-4690 Yang Shen Oliver
Lange Frank Delaglio Paolo Rossi James M.
Aramini Gaohua Liu Alexander Eletsky Yibing
Wu Kiran K. Singarapu Alexander Lemak Alexandr
Ignatchenko Cheryl H. Arrowsmith Thomas
Szyperski Gaetano T. Montelione David Baker Ad Bax
151
The SPARTA Program of Shen and Bax
152
(No Transcript)
153
(No Transcript)
154
Using SPARTA Chemical Shift Prediction to Improve
ROSETTA Scoring Function
155
CS-ROSETTA
156
Performance of CS-ROSETTA on 16 Proteins used for
Optimization
157
(No Transcript)
158
CS-ROSETTA performance on nine structural
genomics proteins
PsR211
NeR45A
StR82
159
(No Transcript)
160
Structures of two designed proteins with
highsequence identity
NMR structures of Ga88 and Gb88
NMR structures vs csRosetta models
Mean-to-mean backbone RMSD 1.31A
1.07A
Patrick A. Alexander, Yanan He, Yihong Chen, John
Orban, and Philip N. Bryan PNAS, 2007,
10411963-11968 PNAS, 2008, 10514412-14417
161
NMR Applications in Drug Discovery
162
Identification of Stereochemistry by Dipolar
Couplings
J.L. Yan, F. Delaglio, A. Kaerner, A.D. Kline,
H.P. Mo, M.J. Shapiro, T.A. Smitka, G.A.
Stephenson, E.R. Zartler Complete relative
stereochemistry of multiple stereocenters using
only residual dipolar couplings. J. Am. Chem.
Soc., 126 (15) 5008-5017 (2004).
163
(No Transcript)
164
(No Transcript)
165
(No Transcript)
166
(No Transcript)
167
NMR Spectral SeriesTwo Approaches
Applications of NMR in the Drug Discovery Process
SAR by NMR (Abbott Labs)
Observe Ligand Signals
Observe Protein Signals
168
(No Transcript)
169
(No Transcript)
170
Amide-N Chemical Shift of Residue i
Amide-HN Chemical Shift of Residue i
171
(No Transcript)
172
(No Transcript)
173
Analyze Titration Curve to Estimate Kd
174
Entire spectrum is a single object in
multdimensional space. Coordinates of the object
are the spectral intensities. Similar spectra
cluster together. Spectra with similar features
lie along lines and curves
175
(No Transcript)
176
(No Transcript)
177
(No Transcript)
178
(No Transcript)
179
(No Transcript)
180
www.nmrscience.com