Characterization of data analysis methods for information recovery from metabolic 1H NMR spectra using artificial complex mixtures

The assessment of data analysis methods in ¹H NMR based metabolic profiling is hampered owing to a lack of knowledge of the exact sample composition. In this study, an artificial complex mixture design comprising two artificially defined groups designated normal and disease, each containing 30 samples, was implemented using 21 metabolites at concentrations typically found in human urine and having a realistic distribution of inter-metabolite correlations. These artificial mixtures were profiled by ¹H NMR spectroscopy and used to assess data analytical methods in the task of differentiating the two conditions. When metabolites were individually quantified, volcano plots provided an excellent method to track the effect size and significance of the change between conditions. Interestingly, the Welch t test detected a similar set of metabolites changing between classes in both quantified and spectral data, suggesting that differential analysis of ¹H NMR spectra using a false discovery rate correction, taking into account fold changes, is a reliable approach to detect differential metabolites in complex mixture studies. Various multivariate regression methods based on partial least squares (PLS) were applied in discriminant analysis mode. The most reliable methods in quantified and spectral ¹H NMR data were PLS and RPLS linear and logistic regression respectively. A jackknife based strategy for variable selection was assessed on both quantified and spectral data and results indicate that it may be possible to improve on the conventional Orthogonal-PLS methodology in terms of accuracy and sensitivity. A key improvement of our approach consists of objective criteria to select significant signals associated with a condition that provides a confidence level on the discoveries made, which can be implemented in metabolic profiling studies.     

An automated device for the preparation of complex reagent mixtures

A microcomputer-controlled device was built that automatically prepares small volumes of mixtures of up to eight reagents. The operation of the system is fast, flexible, and reliable, thus making possible the routine use of experimental protocols that require large numbers of small volume reagent samples, each having a different composition. In particular, the software we developed for this device handles the preparation of three-antibody staining solutions to be used in triple labeling immunofluorescent flow cytometry experiments that involve only two fluorochromes. In this role, the device is known as an “Immunofluorescence Tomograph.”     

Automated approach for quantitative analysis of complex peptide mixtures from tandem mass spectra

To take advantage of the potential quantitative benefits offered by tandem mass spectrometry, we have modified the method in which tandem mass spectrum data are acquired in 'shotgun' proteomic analyses. The proposed method is not data dependent and is based on the sequential isolation and fragmentation of precursor windows (of 10 m/z) within the ion trap until a desired mass range has been covered. We compared the quantitative figures of merit for this method to those for existing strategies by performing an analysis of the soluble fraction of whole-cell lysates from yeast metabolically labeled in vivo with (15)N. To automate this analysis, we modified software (RelEx) previously written in the Yates lab to generate chromatograms directly from tandem mass spectra. These chromatograms showed improvements in signal-to-noise ratio of approximately three- to fivefold over corresponding chromatograms generated from mass spectrometry scans. In addition, to demonstrate the utility of the data-independent acquisition strategy coupled with chromatogram reconstruction from tandem mass spectra, we measured protein expression levels in two developmental stages of Caenorhabditis elegans.     

LC-MS-based method for the qualitative and quantitative analysis of complex lipid mixtures

A simple and robust LC-MS-based methodology for the investigation of lipid mixtures is described, and its application to the analysis of human lipoprotein-associated lipids is demonstrated. After an optional initial fractionation on Silica 60, normal-phase HPLC-MS on a YMC PVA-Sil column is used first for class separation, followed by reversed-phase LC-MS or LC-tandem mass spectrometry using an Atlantis dC18 capillary column, and/or nanospray MS, to fully characterize the individual lipids. The methodology is applied here for the analysis of human apolipoprotein B-associated lipids. This approach allows for the determination of even low percentages of lipids of each molecular species and showed clear differences between lipids associated with apolipoprotein B-100-LDL isolated from a normal individual and those associated with a truncated version, apolipoprotein B-67-containing lipoproteins, isolated from a homozygote patient with familial hypobetalipoproteinemia. The methods described should be easily adaptable to most modern MS instrumentation.     

A general Bayesian method for an automated signal class recognition in 2D NMR spectra combined with a multivariate discriminant analysis

Summary A generally applicable method for the automated classification of 2D NMR peaks has been developed, based on a Bayesian approach coupled to a multivariate linear discriminant analysis of the data. The method can separate true NMR signals from noise signals, solvent stripes and artefact signals. The analysis relies on the assumption that the different signal classes have different distributions of specific properties such as line shapes, line widths and intensities. As to be expected, the correlation network of the distributions of the selected properties affects the choice of the discriminant function and the final selection of signal properties. The classification rule for the signal classes was deduced from Bayes's theorem. The method was successfully tested on a NOESY spectrum of HPr protein from Staphylococcus aureus. The calculated probabilities for the different signal class memberships are realistic and reliable, with a high efficiency of discrimination between peaks that are true NOE signals and those that are not.     

Use of optimized 1D TOCSY NMR for improved quantitation and metabolomic analysis of biofluids

One dimensional selective TOCSY experiments have been shown to be advantageous in providing improved data inputs for principle component analysis (PCA) (Sandusky and Raftery 2005a, b). Better subpopulation cluster resolution in the observed scores plots results from the ability to isolate metabolite signals of interest via the TOCSY based filtering approach. This report reexamines the quantitative aspects of this approach, first by optimizing the 1D TOCSY experiment as it relates to the measurement of biofluid constituent concentrations, and second by comparing the integration of 1D TOCSY read peaks to the bucket integration of 1D proton NMR spectra in terms of precision and accuracy. This comparison indicates that, because of the extensive peak overlap that occurs in the 1D proton NMR spectra of biofluid samples, bucket integrals are often far less accurate as measures of individual constituent concentrations than 1D TOCSY read peaks. Even spectral fitting approaches have proven difficult in the analysis of significantly overlapped spectral regions. Measurements of endogenous taurine made over a sample population of human urine demonstrates that, due to background signals from other constituents, bucket integrals of 1D proton spectra routinely overestimate the taurine concentrations and distort its variation over the sample population. As a result, PCA calculations performed using data matrices incorporating 1D TOCSY determined taurine concentrations produce better scores plot subpopulation cluster resolution.     

2D NMR metabonomic analysis: a novel method for automated peak alignment

MOTIVATION: Comparative metabolic profiling by nuclear magnetic resonance (NMR) is showing increasing promise for identifying inter-individual differences to drug response. Two dimensional (2D) (1)H (13)C NMR can reduce spectral overlap, a common problem of 1D (1)H NMR. However, the peak alignment tools for 1D NMR spectra are not well suited for 2D NMR. An automated and statistically robust method for aligning 2D NMR peaks is required to enable comparative metabonomic analysis using 2D NMR. RESULTS: A novel statistical method was developed to align NMR peaks that represent the same chemical groups across multiple 2D NMR spectra. The degree of local pattern match among peaks in different spectra is assessed using a similarity measure, and a heuristic algorithm maximizes the similarity measure for peaks across the whole spectrum. This peak alignment method was used to align peaks in 2D NMR spectra of endogenous metabolites in liver extracts obtained from four inbred mouse strains in the study of acetaminophen-induced liver toxicity. This automated alignment method was validated by manual examination of the top 50 peaks as ranked by signal intensity. Manual inspection of 1872 peaks in 39 different spectra demonstrated that the automated algorithm correctly aligned 1810 (96.7%) peaks. AVAILABILITY: Algorithm is available upon request.     

An automated device for the preparation of complex reagent mixtures. The immunofluorescence tomograph

A microcomputer-controlled device was built that automatically prepares small volumes of mixtures of up to eight reagents. The operation of the system is fast, flexible, and reliable, thus making possible the routine use of experimental protocols that require large numbers of small volume reagent samples, each having a different composition. In particular, the software we developed for this device handles the preparation of three-antibody staining solutions to be used in triple labeling immunofluorescent flow cytometry experiments that involve only two fluorochromes. In this role, the device is known as an "Immunofluorescence Tomograph."     

Tools for the automated assignment of high-resolution three-dimensional protein NMR spectra based on pattern recognition techniques

One of the major bottlenecks in the determination of proteinstructures by NMR is in the evaluation of the data produced by theexperiments. An important step in this process is assignment, where thepeaks in the spectra are assigned to specific spins within specificresidues. In this paper, we discuss a spin system assignment tool based onpattern recognition techniques. This tool employs user-specified templatesto search for patterns of peaks in the original spectra; these patterns maycorrespond to side-chain or backbone fragments. Multiple spectra willnormally be searched simultaneously to reduce the impact of noise. Thesearch generates a preliminary list of putative assignments, which arefiltered by a set of heuristic algorithms to produce the final results list.Each result contains a set of chemical shift values plus information aboutthe peaks found. The results may be used as input for combinatorialroutines, such as sequential assignment procedures, in place of peak lists.Two examples are presented, in which (i) HCCH-COSY and -TOCSY spectra arescanned for side-chain spin systems; and (ii) backbone spin systems aredetected in a set of spectra comprising HNCA, HN(CO)CA, HNCO, HN(CA)CO,CBCANH and CBCA(CO)NH.     

Application of neural networks to automated assignment of NMR spectra of proteins

Summary Simulated neural networks are described which aid the assignment of protein NMR spectra. A network trained to recognize amino acid type from TOCSY data was trained on 148 assigned spin systems from E. coli acyl carrier proteins (ACPs) and tested on spin systems from spinach ACP, which has a 37% sequence homology with E. coli ACP and a similar secondary structure. The output unit corresponding to the correct amino acid is one of the four most activated units in 83% of the spin systems tested. The utility of this information is illustrated by a second network which uses a constraint satisfaction algorithm to find the best fit of the spin systems to the amino acid sequence. Application to a stretch of 20 amino acids in spinach ACP results in 75% correct sequential assignment. Since the output of the amino acid type identification network can be coupled with a variety of sequential assignment strategies, the approach offers substantial potential for expediting assignment of protein NMR spectra.     

NMR spectroscopic filtration of polypeptides and proteins in complex mixtures

Due to the inherent complexity of the natural biological environment, most studies on polypeptides, proteins and nucleic acids have so far been performed in vitro, away from physiologically relevant conditions. Nuclear magnetic resonance is an ideal technique to extend the in vitro analysis of simple model systems to the more complex biological context. This work shows how diffusion-based spectroscopic selection can be combined with isotopic labeling to tackle and optimize the NMR analysis of specific macromolecules in multicomponent mixtures. Typical media include cell-free systems containing overexpressed proteins, lysates and proteolytic mixtures. We present a few variants of diffusion-edited HSQC pulse sequences for the selective spectroscopic detection of protein and polypeptide resonances within complex mixtures containing undesired species of smaller molecular weight. Due to diffusion-based filtering, peak intensities of fast diffusing small molecules are attenuated more than peaks due to large molecules. The basic sequence, denoted as PFGSTE-HSQC, combines translational diffusion-ordering with two dimensional heteronuclear single quantum correlation spectroscopy. The GCSTE-HSQC and BPPSTE-HSQC sequences include bipolar gradients and are therefore suitable for both diffusion-based filtering and determination of diffusion coefficients of individual mixture components. Practical applications range from protein stability/folding investigations in physiologically relevant contexts to prescreening of tertiary fold and resonance assignments in structural genomics studies. A few applications of diffusion-edited HSQC to an E. coli cell lysate containing the (15)N-labeled B domain of streptococcal protein G (GB1), and to a (15)N-labeled N-acetylglycine/apomyoglobin mixture are presented. In addition, we provide specific guidelines for experimental setup and parameter optimization.     

Automated screening for metabolites in complex mixtures using 2D COSY NMR spectroscopy

One of the greatest challenges in metabolomics is the rapid and unambiguous identification and quantification of metabolites in a biological sample. Although one-dimensional (1D) proton nuclear magnetic resonance (NMR) spectra can be acquired rapidly, they are complicated by severe peak overlap that can significantly hinder the automated identification and quantification of metabolites. Furthermore, it is currently not reasonable to assume that NMR spectra of pure metabolites are available a priori for every metabolite in a biological sample. In this paper we develop and report on tests of methods that assist in the automatic identification of metabolites using proton two-dimensional (2D) correlation spectroscopy (COSY) NMR. Given a database of 2D COSY spectra for the metabolites of interest, our methods provide a list sorted by a heuristic likelihood of the metabolites present in a sample that has been analyzed using 2D COSY NMR. Our models attempt to correct the displacement of the peaks that can occur from one sample to the next, due to pH, temperature and matrix effects, using a statistical and chemical model. The correction of one peak can result in an implied correction of others due to spin–spin coupling. Furthermore, these displacements are not independent: they depend on the relative position of functional groups in the molecule. We report experimental results using defined mixtures of amino acids as well as real complex biological samples that demonstrate that our methods can be very effective at automatically and rapidly identifying metabolites.     

AURELIA,a program for computer-aided analysis of multidimensional NMR spectra

Summary AURELIA is an advanced program for the computer-aided evaluation of two-, three- and four-dimensional NMR spectra of any type of molecule. It can be used for the analysis of spectra of small molecules as well as for evaluation of complicated spectra of biological macromolecules such as proteins. AURELIA is highly interactive and offers a large number of tools, such as artefact reduction, cluster and multiplet analysis, spin system searches, resonance assignments, automated calculation of volumes in multidimensional spectra, calculation of distances with different approaches, including the full relaxation matrix approach, Bayesian analysis of peak features, correlation of molecular structures with NMR data, comparison of spectra via spectral algebra and pattern match techniques, automated sequential assignments on the basis of triple resonance spectra, and automatic strip calculation. In contrast to most other programs, many tasks are performed automatically.