HIFI NMR : part1 automated backbone assignments using 3D->2D

1. HIFI NMR : part1automated backbone assignments using 3D->2D Marco Tonelli National Magnetic Resonance Facility At Madison NMRFAM

2. t1 t t1 t2 t2 t3 t1x t2: 128x128=16,384 FIDs 5 days 16 hrs. 1 FID 30 sec. t1: 128 FIDs 64 min. 1D 2D 3D t1x t2 x t3 : 32x32x32=32,768 FIDs 11days 9 hrs. 4D Recording multidimensional experiments is time costly In conventional multidimensional experiments, all the individual frequency domains are incremented (sampled) independently • Increasing number of indirect dimensions in conventional multidimensional: • collection time increases exponentially • need to reduce number of increments to keep collection time reasonable: low resolution • not very practical above 3 dimensions • impossible above 4 dimensions

3. Fast methods Reduced Sampling Reduced Dimensionality Hadamard spectroscopy single-scan NMR so-fast NMR

4. Conventional 3D spectrum RD spectrum t1/t2 t3 t1 t2 t1 and t2 are incremented simultaneously point by point t3 t1 and t2 are incremented independently Reduced Dimensionality Techniques two or more indirect dimensions are evolved simultaneously t3 t1 t2 preparation mix/prep mixing

5. Reduced Dimensionality Techniques … F1 F2 t3 t1 t2 preparation mix/prep mixing F1 = x,y F2 = y F1 = x,y F2 = x t1/t2 t1/t2 RD spectrum RD spectrum F1 = x,y sinwt1 x sinwt2 sinwt1 x coswt2 coswt1 x sinwt2 coswt1 x coswt2 t1 2D spectrum t3 t3 FT FT sinwt1 coswt1 d1-d2 d1- d2 t3 w1/w2 FT w1/w2 d1+d2 d1+d2 w3 w3 d1 w1 d1-d2 w3 w1/w2 d1+d2 w3 w3

6. Reduced Dimensionality Techniques … GFT TPPI Method d1+ d2 + d2 d1+ d2 2D w1/w2 TPPI offset w1 d1 - d2 d1- d2 d1- d2 w3 w3 The peaks in RD experiments can either be separated into different spectra (GFT - Szyperski) or into different regions of the same spectrum (TPPI - Gronenborn)

7. 0° 1H-13C plane of CBCA(CO)NH 13C 0° plane 1H 90° 15N 13C 1H-15N plane of CBCA(CO)NH 15N 90° plane 15N 1H 13C 1H 1H q 15N 45° 1H-13C/15N plane of CBCA(CO)NH 13C/15N tilted plane 13C 1H 1H 2D RD planes of 3D spectra 2D projections of 3D spectra tilted planes

8. Reduced Dimensionality Techniques … t 13C 15N t 15N 13C 1H By changing the ratio between the two simultaneously evolving dimensions we can change the projection angle of the tilted plane q 20° 70° 45° q 2D planes of 3D CBCA(CO)NH

9. Reduced Dimensionality Techniques … 3D reconstruction 3D object reconstructed object 15N 13C 90 1H 3D reconstrucion 0 q  Projection Reconstruction Simple 3D objects can be reconstructed from 2D projections collected at different angles In principle, it is feasible to reconstruct a 3D spectrum from a number of 2D tilted planes collected at different angles.

10. Reduced Dimensionality Techniques … n-dimensional experiments are run as two-dimensional RD spectra multiple tilted planes are collected at different angles n-dimensional spectra are reconstructed from several tilted planes split peaks can be separated into different spectra (GFT) or different regions of the same spectrum (TPPI) indirect frequencies are extracted from analysis of split patterns GFT / TPPI method Projection-Reconstruction High-resolution Iterative Frequency Identification HIFI simultaneously evolving indirect frequencies are extracted from two-dimensional RD spectra multiple tilted planes are used angle of tilted planes is chosen adaptively in real time

11. Reduced Dimensionality Techniques … it relies on combining multiple indirect dimension to resolve overlapped peaks multidimensional frequency information can only be extracted after spectra reconstruction • with each added indirect dimension: • S/N is reduced by √2 • collection time is doubled reconstructing spectra can be very complicated or impossible with overlapped peaks and/or low S/N analysis can be very complicated difficult to automate difficult to automate by changing the tilt angle, HIFI has greater potential of resolving overlapped peaks than GFT/TPPI methods since only frequency information is extracted from tilted planes, HIFI does not need to resolve the problem of reconstructing nD volumes easier to analyze easier to automate by adaptively choosing the next tilt angle, HIFI optimizes information gain while minimizing time collection GFT / TPPI method Projection-Reconstruction HIFI

12. predict next best tilted angle does tilted plane add new information ??? YES this process can be automated in vnmr NO HIFI flowchart mORTHO0 macro orthogonal planes hifi.sh tilted plane generate_peaklist.sh peak list

13. predicted chemical shift distribution orthogonal planes assign a probability of a peak being in a given voxel, p 0° 90° dispersion function, fq(p),measures the dispersion of the putative peaks on the selected tilted plane find a tilt angle that maximizes a dispersion function fq (p) collect tilted plane does tilted plane add new information ??? X° NO YES peak list HIFI algorithm for predicting best tilted angle Eghbalnia et al JACS 127 (36), 12528 -12536, 2005

14. putative peaks from C-H/N-H planes add 45° tilted plane add additional tilted planes HIFI on CBCA(CO)NH Combined peaks from HIFI planes are in magenta Hand picked peaks from 3D spectrum in green

15. Modified BioPack experiments for backbone assignments HNCO NH sensitivity enhanced TROSY option HN(CO)CA HNCA CBCA(CO)NH HN(CA)CB HNCACB standard sequences are robust and offer the best S/N for backbone assignments we rely on HIFI ability to adaptively select tilted planes to resolve overlapped peaks • added HIFI option for collecting tilted planes • 13C and 15N indirect dimensions are evolved simultaneously • added semi-constant time 15N evolution to allow collecting tilted planes with more indirect points for higher resolution • optimized for cryogenic probes Using HIFI for backbone assignments Needs AUTOMATION !!!

16. outline of vnmr macro for automated HIFI data collection preparation - orthogonal planes • run othogonal planes for all experiments • all experiments list: • collect 0º plane • adjust 13C s.w. • 1H-15N HSQC : • use as 90º plane • adjust 15N s.w. input list of experiments number of residues = XX HNCO HN(CO)CA HNCA CBCA(CO)NH HNCACB HN(CA)CB hnco_tilt_0 hncoca_tilt_0 hnca_tilt_0 cbcaconh_tilt_0 hncacb_tilt_0 hncb_tilt_0 hnco_hsqc hnco_hsqc hnco_hsqc hnco_hsqc hnco_hsqc hnco_hsqc • save orthogonal planes • process orthogonal planes • optimize processing parameters experiment #1 in list run HIFI macro - tilted planes run tilt angle predicting program process tilted plane save tilted plane read predicted tilt angle tilt>0 run experiment to collect tilted plane at suggested angle next experiment in list tilt=0 run program to extract 3D frequencies 3D peak list run probabilistic backbone assignments all experiments in list completed ?? NO YES

17. 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 90º 63º 62º 64º 51º 47º 43º 41º 42º 38º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 0º 34º 20º 2 2 4 3 77º 69º 70º 69º 45º 52º 53º 44º 34º 26º 15º 2 2 3 4 71º 71º 63º 59º 67º 52º 41º 51º 46º 56º 39º 38º 33º 40º 34º 27º 29º 22º 18º 7 4 4 4 70º 63º 77º 60º 57º 65º 49º 53º 48º 46º 54º 44º 40º 35º 42º 32º 26º 31º 18º 4 4 7 4 70º 59º 60º 62º 53º 42º 53º 41º 35º 31º 41º 28º 28º 14º 11º 4 4 7 57º 45º 42º 39º 45º 31º 39º 33º 40º 35º 31º 24º 37º 24º 29º 12º 18º 26º 18º 14º 15º 5º 5 5 11º 8º 6 9 3º ~9hrs ~48hrs ~12hrs ~14hrs brazzein - 53 a.a. ubiquitin 76 a.a. flavodoxin 176 a.a. wild-type RI mutant HNCO HN(CO)CA HNCA CBCA(CO)NH HN(CA)CB HNCACB

18. HIFI backbone assignments - recap • we have developed HIFI for extracting 3D peaks using 2D tilted planes • by adaptively predicting the best tilt angles, we guarantee that all available 3D data is extracted using the minimum number of tilted planes • we have adapted the most robust 3D experiments for backbone assignments • to be recorded using HIFI-NMR • we have successfully automated HIFI backbone data collection for proteins of small/medium size • automated HIFI is robust and allows to extract 3D peaks lists with: • least amount of spectrometer time • minimum human intervention

19. longer time≡higher S/N longer time≡higher S/N resolve overlap resolve overlap complicated complicated lower S/N HIFI 4 3 TROSY 6 5 4 conventional NMR experiments 3 2 protein sample 3 size Total # dim Acquired # dim Acquired # dim where is HIFI going … deuteration selective labeling peak overlap T2 relaxation Improving algorithm for peak detection Improving algorithm for tilt angle prediction Combine information from different experiments

20. NMR data collection structure calculation chemical shift assignments The BIG picture

21. HIFI NMR : part2conclusions and other applications Marco Tonelli National Magnetic Resonance Facility At Madison NMRFAM

22. robust : collect all the information needed HIFI adaptive efficient : collect only the minimum amount information needed to answer the question HIFI orthogonal plane predict best angle collect tilted plane peak list HIFI BACKBONE ASSIGNMENTS orthogonal plane predict best angle collect tilted plane HIFI: application to chemical shift assignments TODAY: HNCO HN(CO)CA HNCA CBCA(CO)NH HNCACB HN(CA)CB

23. robust : collect all the information needed HIFI adaptive efficient : collect only the minimum amount information needed to answer the question orthogonal plane predict best angle collect tilted plane peak list BACKBONE ASSIGNMENTS HIFI orthogonal plane predict best angle collect tilted plane peak list HIFI: application to chemical shift assignments TOMORROW: HNCO HN(CO)CA HNCA CBCA(CO)NH HNCACB HN(CA)CB

24. robust : collect all the information needed HIFI adaptive efficient : collect the only the minimum amount information needed to answer the question HIFI collect preliminary data select experiment predict best tilt collect tilted plane BACKBONE ASSIGNMENTS HIFI: application to chemical shift assignments THE DAY AFTER TOMORROW: pool of experiments

25. Conventional 3D spectrum tilted plane t1/t2 t3 t1 t2 t3 HIFI: other applications any application that makes use of information extracted from a 3D experiment can be speeded up by recording a tilted plane instead HIFI can then be used to ensure that maximum information is recovered by predicting the best angle to use for recording the tilted plane

26. +q 13C + 15N 1H 13C - 15N -q 1H 15N 13C 1H HIFI: other applications • two tilted planes are obtained for each experiment run: “plus” and “minus” • different peak distribution – bigger potential of resolving overlapped peaks • different noise distribution – can provide measure of confidence of data obtained by analyzing spectra

27. attenuated reference 15N 15N 13C 13C 1H 1H HIFI: extraction of RDC Example: extraction of RDCC’N using a modified HNCO pulse sequence (Bax) Conventional method: • record two 3D C’-N-H experiments: • reference spectrum • attenuated spectrum - intensity of peaks is modulated by coupling: JC’N + DC’N • extract JC’N + DC’N coupling from the ratio between intensity of corresponding peaks in the reference and attenuated spectra • repeat for isotropic and aligned samples

28. +q reference attenuated reference attenuated -q +q 13C + 15N 1H 1H -q 13C - 15N 1H 1H HIFI: extraction of RDC HIFI method: • record two tilted C’-N-H planes at the optimal tilt angle: • reference spectrum • attenuated spectrum - intensity of peaks is modulated by coupling: JC’N + DC’N • extract JC’N + DC’N coupling from the ratio between intensity of corresponding peaks in the reference and attenuated spectra • analyze “plus” and “minus” planes independently • compare the results from the two plenes to get measure of data confidence • combine the results from the two planes • repeat for isotropic and aligned samples

29. tilt prediction and peak extraction algorithms Arash Barhami Hamid Eghbalnia pulse sequences vnmr automation RDC extraction Marco Tonelli Klaas Hallenga Gabriel Cornilescu sample preparation Fariba Assadi-Porter Shanteri Shingh Rob Tyler Anna Fuzery Nick Reiter Claudia Cornilescu Who did the work ? John Markley Milo Westler

30. 600MHz 750MHz 900MHz 800MHz NMRFAM Thank you !