CheckShift: automatic correction of inconsistent chemical shift referencing

Automatic 13C chemical shift reference correction for unassigned protein NMR spectra

Poor chemical shift referencing, especially for ¹³C in protein Nuclear Magnetic Resonance (NMR) experiments, fundamentally limits and even prevents effective study of biomacromolecules via NMR, including protein structure determination and analysis of protein dynamics. To solve this problem, we constructed a Bayesian probabilistic framework that circumvents the limitations of previous reference correction methods that required protein resonance assignment and/or three-dimensional protein structure. Our algorithm named Bayesian Model Optimized Reference Correction (BaMORC) can detect and correct ¹³C chemical shift referencing errors before the protein resonance assignment step of analysis and without three-dimensional structure. By combining the BaMORC methodology with a new intra-peaklist grouping algorithm, we created a combined method called Unassigned BaMORC that utilizes only unassigned experimental peak lists and the amino acid sequence. Unassigned BaMORC kept all experimental three-dimensional HN(CO)CACB-type peak lists tested within ± 0.4 ppm of the correct ¹³C reference value. On a much larger unassigned chemical shift test set, the base method kept ¹³C chemical shift referencing errors to within ± 0.45 ppm at a 90% confidence interval. With chemical shift assignments, Assigned BaMORC can detect and correct ¹³C chemical shift referencing errors to within ± 0.22 at a 90% confidence interval. Therefore, Unassigned BaMORC can correct ¹³C chemical shift referencing errors when it will have the most impact, right before protein resonance assignment and other downstream analyses are started. After assignment, chemical shift reference correction can be further refined with Assigned BaMORC. These new methods will allow non-NMR experts to detect and correct ¹³C referencing error at critical early data analysis steps, lowering the bar of NMR expertise required for effective protein NMR analysis.

NMR assignments of the N-terminal signaling domain of the TonB-dependent outer membrane transducer PupB

Outer membrane TonB-dependent transducers (TBDTs) actively transport ferric siderophore complexes from the extracellular environment into Gram-negative bacteria. They also participate in a cell-surface signaling regulatory pathway that results in upregulation of the transducer itself, in response to iron-deplete conditions. The TBDT PupB transports ferric pseudobactin, and signals through its N-terminal signaling domain (NTSD), while the TBDT homolog PupA is signaling-inactive. Here, we report the NMR chemical shift assignments of the PupB-NTSD. This information will provide the basis for structural characterization of the PupB-NTSD to further explore its signaling properties.

Conformationally selective multidimensional chemical shift ranges in proteins from a PACSY database purged using intrinsic quality criteria

We have determined refined multidimensional chemical shift ranges for intra-residue correlations ((13)C-(13)C, (15)N-(13)C, etc.) in proteins, which can be used to gain type-assignment and/or secondary-structure information from experimental NMR spectra. The chemical-shift ranges are the result of a statistical analysis of the PACSY database of >3000 proteins with 3D structures (1,200,207 (13)C chemical shifts and >3 million chemical shifts in total); these data were originally derived from the Biological Magnetic Resonance Data Bank. Using relatively simple non-parametric statistics to find peak maxima in the distributions of helix, sheet, coil and turn chemical shifts, and without the use of limited "hand-picked" data sets, we show that ~94% of the (13)C NMR data and almost all (15)N data are quite accurately referenced and assigned, with smaller standard deviations (0.2 and 0.8 ppm, respectively) than recognized previously. On the other hand, approximately 6% of the (13)C chemical shift data in the PACSY database are shown to be clearly misreferenced, mostly by ca. -2.4 ppm. The removal of the misreferenced data and other outliers by this purging by intrinsic quality criteria (PIQC) allows for reliable identification of secondary maxima in the two-dimensional chemical-shift distributions already pre-separated by secondary structure. We demonstrate that some of these correspond to specific regions in the Ramachandran plot, including left-handed helix dihedral angles, reflect unusual hydrogen bonding, or are due to the influence of a following proline residue. With appropriate smoothing, significantly more tightly defined chemical shift ranges are obtained for each amino acid type in the different secondary structures. These chemical shift ranges, which may be defined at any statistical threshold, can be used for amino-acid type assignment and secondary-structure analysis of chemical shifts from intra-residue cross peaks by inspection or by using a provided command-line Python script (PLUQin), which should be useful in protein structure determination. The refined chemical shift distributions are utilized in a simple quality test (SQAT) that should be applied to new protein NMR data before deposition in a databank, and they could benefit many other chemical-shift based tools.

NMR-based modeling and refinement of protein 3D structures

NMR is a well-established method to characterize the structure and dynamics of biomolecules in solution. High-quality structures can now be produced thanks to both experimental advances and computational developments that incorporate new NMR parameters and improved protocols and force fields in the structure calculation and refinement process. In this chapter, we give a short overview of the various types of NMR data that can provide structural information, and then focus on the structure calculation methodology itself. We discuss and illustrate with tutorial examples "classical" structure calculation, refinement, and structure validation approaches.

An overview of tools for the validation of protein NMR structures

Biomolecular structures at atomic resolution present a valuable resource for the understanding of biology. NMR spectroscopy accounts for 11% of all structures in the PDB repository. In response to serious problems with the accuracy of some of the NMR-derived structures and in order to facilitate proper analysis of the experimental models, a number of program suites are available. We discuss nine of these tools in this review: PROCHECK-NMR, PSVS, GLM-RMSD, CING, Molprobity, Vivaldi, ResProx, NMR constraints analyzer and QMEAN. We evaluate these programs for their ability to assess the structural quality, restraints and their violations, chemical shifts, peaks and the handling of multi-model NMR ensembles. We document both the input required by the programs and output they generate. To discuss their relative merits we have applied the tools to two representative examples from the PDB: a small, globular monomeric protein (Staphylococcal nuclease from S. aureus, PDB entry 2kq3) and a small, symmetric homodimeric protein (a region of human myosin-X, PDB entry 2lw9).

Recommendations of the wwPDB NMR Validation Task Force

As methods for analysis of biomolecular structure and dynamics using nuclear magnetic resonance spectroscopy (NMR) continue to advance, the resulting 3D structures, chemical shifts, and other NMR data are broadly impacting biology, chemistry, and medicine. Structure model assessment is a critical area of NMR methods development, and is an essential component of the process of making these structures accessible and useful to the wider scientific community. For these reasons, the Worldwide Protein Data Bank (wwPDB) has convened an NMR Validation Task Force (NMR-VTF) to work with wwPDB partners in developing metrics and policies for biomolecular NMR data harvesting, structure representation, and structure quality assessment. This paper summarizes the recommendations of the NMR-VTF, and lays the groundwork for future work in developing standards and metrics for biomolecular NMR structure quality assessment.

NightShift: NMR shift inference by general hybrid model training--a framework for NMR chemical shift prediction

Background

NMR chemical shift prediction plays an important role in various applications in computational biology. Among others, structure determination, structure optimization, and the scoring of docking results can profit from efficient and accurate chemical shift estimation from a three-dimensional model.

A variety of NMR chemical shift prediction approaches have been presented in the past, but nearly all of these rely on laborious manual data set preparation and the training itself is not automatized, making retraining the model, e.g., if new data is made available, or testing new models a time-consuming manual chore.

Results

In this work, we present the framework NightShift (NMR Shift Inference by General Hybrid Model Training), which enables automated data set generation as well as model training and evaluation of protein NMR chemical shift prediction.

In addition to this main result – the NightShift framework itself – we describe the resulting, automatically generated, data set and, as a proof-of-concept, a random forest model called Spinster that was built using the pipeline.

Conclusion

By demonstrating that the performance of the automatically generated predictors is at least en par with the state of the art, we conclude that automated data set and predictor generation is well-suited for the design of NMR chemical shift estimators.

The framework can be downloaded from https://bitbucket.org/akdehof/nightshift. It requires the open source Biochemical Algorithms Library (BALL), and is available under the conditions of the GNU Lesser General Public License (LGPL). We additionally offer a browser-based user interface to our NightShift instance employing the Galaxy framework via https://ballaxy.bioinf.uni-sb.de/.

Peakmatch: a simple and robust method for peak list matching

Peak lists are commonly used in NMR as input data for various software tools such as automatic assignment and structure calculation programs. Inconsistencies of chemical shift referencing among different peak lists or between peak and chemical shift lists can cause severe problems during peak assignment. Here we present a simple and robust tool to achieve self-consistency of the chemical shift referencing among a set of peak lists. The Peakmatch algorithm matches a set of peak lists to a specified reference peak list, neither of which have to be assigned. The chemical shift referencing offset between two peak lists is determined by optimizing an assignment-free match score function using either a complete grid search or downhill simplex optimization. It is shown that peak lists from many different types of spectra can be matched reliably as long as they contain at least two corresponding dimensions. Using a simulated peak list, the Peakmatch algorithm can also be used to obtain the optimal agreement between a chemical shift list and experimental peak lists. Combining these features makes Peakmatch a useful tool that can be applied routinely before automatic assignment or structure calculation in order to obtain an optimized input data set.

Functional binding of hexanucleotides to 3C protease of hepatitis A virus

Oligonucleotides as short as 6 nt in length have been shown to bind specifically and tightly to proteins and affect their biological function. Yet, sparse structural data are available for corresponding complexes. Employing a recently developed hexanucleotide array, we identified hexadeoxyribonucleotides that bind specifically to the 3C protease of hepatitis A virus (HAV 3C(pro)). Inhibition assays in vitro identified the hexanucleotide 5'-GGGGGT-3' (G(5)T) as a 3C(pro) protease inhibitor. Using (1)H NMR spectroscopy, G(5)T was found to form a G-quadruplex, which might be considered as a minimal aptamer. With the help of (1)H, (15)N-HSQC experiments the binding site for G(5)T was located to the C-terminal β-barrel of HAV 3C(pro). Importantly, the highly conserved KFRDI motif, which has previously been identified as putative viral RNA binding site, is not part of the G(5)T-binding site, nor does G(5)T interfere with the binding of viral RNA. Our findings demonstrate that sequence-specific nucleic acid-protein interactions occur with oligonucleotides as small as hexanucleotides and suggest that these compounds may be of pharmaceutical relevance.

A procedure to validate and correct the 13C chemical shift calibration of RNA datasets

Chemical shifts reflect the structural environment of a certain nucleus and can be used to extract structural and dynamic information. Proper calibration is indispensable to extract such information from chemical shifts. Whereas a variety of procedures exist to verify the chemical shift calibration for proteins, no such procedure is available for RNAs to date. We present here a procedure to analyze and correct the calibration of (13)C NMR data of RNAs. Our procedure uses five (13)C chemical shifts as a reference, each of them found in a narrow shift range in most datasets deposited in the Biological Magnetic Resonance Bank. In 49 datasets we could evaluate the (13)C calibration and detect errors or inconsistencies in RNA (13)C chemical shifts based on these chemical shift reference values. More than half of the datasets (27 out of those 49) were found to be improperly referenced or contained inconsistencies. This large inconsistency rate possibly explains that no clear structure-(13)C chemical shift relationship has emerged for RNA so far. We were able to recalibrate or correct 17 datasets resulting in 39 usable (13)C datasets. 6 new datasets from our lab were used to verify our method increasing the database to 45 usable datasets. We can now search for structure-chemical shift relationships with this improved list of (13)C chemical shift data. This is demonstrated by a clear relationship between ribose (13)C shifts and the sugar pucker, which can be used to predict a C2'- or C3'-endo conformation of the ribose with high accuracy. The improved quality of the chemical shift data allows statistical analysis with the potential to facilitate assignment procedures, and the extraction of restraints for structure calculations of RNA.

Structural origins of high apparent dielectric constants experienced by ionizable groups in the hydrophobic core of a protein

The side chains of Lys66, Asp66, and Glu66 in staphylococcal nuclease are fully buried and surrounded mainly by hydrophobic matter, except for internal water molecules associated with carboxylic oxygen atoms. These ionizable side chains titrate with pK(a) values of 5.7, 8.8, and 8.9, respectively. To reproduce these pK(a) values with continuum electrostatics calculations, we treated the protein with high dielectric constants. We have examined the structural origins of these high apparent dielectric constants by using NMR spectroscopy to characterize the structural response to the ionization of these internal side chains. Substitution of Val66 with Lys66 and Asp66 led to increased conformational fluctuations of the microenvironments surrounding these groups, even under pH conditions where Lys66 and Asp66 are neutral. When Lys66, Asp66, and Glu66 are charged, the proteins remain almost fully folded, but resonances for a few backbone amides adjacent to the internal ionizable residues are broadened. This suggests that the ionization of the internal groups promotes a local increase in dynamics on the intermediate timescale, consistent with either partial unfolding or increased backbone fluctuations of helix 1 near residue 66, or, less likely, with increased fluctuations of the charged side chains at position 66. These experiments confirm that the high apparent dielectric constants reported by internal Lys66, Asp66, and Glu66 reflect localized changes in conformational fluctuations without incurring detectable global structural reorganization. To improve structure-based pK(a) calculations in proteins, we will need to learn how to treat this coupling between ionization of internal groups and local changes in conformational fluctuations explicitly.

Validation of archived chemical shifts through atomic coordinates

The public archives containing protein information in the form of NMR chemical shift data at the BioMagResBank (BMRB) and of 3D structure coordinates at the Protein Data Bank are continuously expanding. The quality of the data contained in these archives, however, varies. The main issue for chemical shift values is that they are determined relative to a reference frequency. When this reference frequency is set incorrectly, all related chemical shift values are systematically offset. Such wrongly referenced chemical shift values, as well as other problems such as chemical shift values that are assigned to the wrong atom, are not easily distinguished from correct values and effectively reduce the usefulness of the archive. We describe a new method to correct and validate protein chemical shift values in relation to their 3D structure coordinates. This method classifies atoms using two parameters: the per-atom solvent accessible surface area (as calculated from the coordinates) and the secondary structure of the parent amino acid. Through the use of Gaussian statistics based on a large database of 3220 BMRB entries, we obtain per-entry chemical shift corrections as well as Z scores for the individual chemical shift values. In addition, information on the error of the correction value itself is available, and the method can retain only dependable correction values. We provide an online resource with chemical shift, atom exposure, and secondary structure information for all relevant BMRB entries (http://www.ebi.ac.uk/pdbe/nmr/vasco) and hope this data will aid the development of new chemical shift-based methods in NMR. Proteins 2010. © 2010 Wiley-Liss, Inc.

Backbone and side chain 1H, 15N and 13C assignments for a thiol-disulphide oxidoreductase from the Antarctic bacterium Pseudoalteromonas haloplanktis TAC125

Enzymes produced by psychrophilic organisms have successfully overcome the low temperature challenge and evolved to maintain high catalytic rates in their permanently cold environments. As an initial step in our attempt to elucidate the cold-adaptation strategies used by these enzymes we report here the (1)H, (15)N and (13)C assignments for the reduced form of a thiol-disulphide oxidoreductase from the Antarctic bacterium Pseudoalteromonas haloplanktis TAC125.

A probabilistic approach for validating protein NMR chemical shift assignments

It has been estimated that more than 20% of the proteins in the BMRB are improperly referenced and that about 1% of all chemical shift assignments are mis-assigned. These statistics also reflect the likelihood that any newly assigned protein will have shift assignment or shift referencing errors. The relatively high frequency of these errors continues to be a concern for the biomolecular NMR community. While several programs do exist to detect and/or correct chemical shift mis-referencing or chemical shift mis-assignments, most can only do one, or the other. The one program (SHIFTCOR) that is capable of handling both chemical shift mis-referencing and mis-assignments, requires the 3D structure coordinates of the target protein. Given that chemical shift mis-assignments and chemical shift re-referencing issues should ideally be addressed prior to 3D structure determination, there is a clear need to develop a structure-independent approach. Here, we present a new structure-independent protocol, which is based on using residue-specific and secondary structure-specific chemical shift distributions calculated over small (3-6 residue) fragments to identify mis-assigned resonances. The method is also able to identify and re-reference mis-referenced chemical shift assignments. Comparisons against existing re-referencing or mis-assignment detection programs show that the method is as good or superior to existing approaches. The protocol described here has been implemented into a freely available Java program called "Probabilistic Approach for protein Nmr Assignment Validation (PANAV)" and as a web server ( http://redpoll.pharmacy.ualberta.ca/PANAV ) which can be used to validate and/or correct as well as re-reference assigned protein chemical shifts.

CheckShift improved: fast chemical shift reference correction with high accuracy

The construction of a consistent protein chemical shift database is an important step toward making more extensive use of this data in structural studies. Unfortunately, progress in this direction has been hampered by the quality of the available data, particularly with respect to chemical shift referencing, which is often either inaccurate or inconsistently annotated. Preprocessing of the data is therefore required to detect and correct referencing errors. In an earlier study we developed CheckShift, a program for performing this task automatically. Now we spent substantial effort in improving the running time of the CheckShift algorithm, which resulted in an running time decrease of 90%, thereby achieving equivalent quality to the former version of CheckShift. The reason for the running time decrease is twofold. Firstly we improved the search for the optimal re-referencing offset considerably. Secondly, as CheckShift is based on a secondary structure prediction from the amino acid sequence (formally PsiPred was used), we evaluated a wide range of available secondary structure prediction programs focusing on the special needs of the CheckShift algorithm. The results of this evaluation prove empirically that we can use faster secondary structure prediction programs than PsiPred without sacrificing CheckShift's accuracy. Very recently Wang and Markley (2009) gave a small list of extreme outliers of the former version of the CheckShift web-server. Those were due to the empirical reduction of the search space implemented in the old version. The new version of CheckShift now gives very similar results to RefDB and LACS for all outliers mentioned in Table 1 of Wang and Markley (2009).

Empirical correlation between protein backbone 15N and 13C secondary chemical shifts and its application to nitrogen chemical shift re-referencing

The linear analysis of chemical shifts (LACS) has provided a robust method for identifying and correcting 13C chemical shift referencing problems in data from protein NMR spectroscopy. Unlike other approaches, LACS does not require prior knowledge of the three-dimensional structure or inference of the secondary structure of the protein. It also does not require extensive assignment of the NMR data. We report here a way of extending the LACS approach to 15N NMR data from proteins, so as to enable the detection and correction of inconsistencies in chemical shift referencing for this nucleus. The approach is based on our finding that the secondary 15N chemical shift of the backbone nitrogen atom of residue i is strongly correlated with the secondary chemical shift difference (experimental minus random coil) between the alpha and beta carbons of residue i-1. Thus once alpha and beta 13C chemical shifts are available (their difference is referencing error-free), the 15N referencing can be validated, and an appropriate offset correction can be derived. This approach can be implemented prior to a structure determination and can be used to analyze potential referencing problems in database data not associated with three-dimensional structure. Application of the LACS algorithm to the current BMRB protein chemical shift database, revealed that nearly 35% of the BMRB entries have delta 15N values mis-referenced by over 0.7 ppm and over 25% of them have delta 1HN values mis-referenced by over 0.12 ppm. One implication of the findings reported here is that a backbone 15N chemical shift provides a better indicator of the conformation of the preceding residue than of the residue itself.

SimShiftDB; local conformational restraints derived from chemical shift similarity searches on a large synthetic database

We present SimShiftDB, a new program to extract conformational data from protein chemical shifts using structural alignments. The alignments are obtained in searches of a large database containing 13,000 structures and corresponding back-calculated chemical shifts. SimShiftDB makes use of chemical shift data to provide accurate results even in the case of low sequence similarity, and with even coverage of the conformational search space. We compare SimShiftDB to HHSearch, a state-of-the-art sequence-based search tool, and to TALOS, the current standard tool for the task. We show that for a significant fraction of the predicted similarities, SimShiftDB outperforms the other two methods. Particularly, the high coverage afforded by the larger database often allows predictions to be made for residues not involved in canonical secondary structure, where TALOS predictions are both less frequent and more error prone. Thus SimShiftDB can be seen as a complement to currently available methods.