MOTIVATION: Metabolomics extensively utilizes nuclear magnetic resonance (NMR) spectroscopy due to its excellent reproducibility and high throughput. Both 1D and 2D NMR spectra provide crucial information for metabolite annotation and quantification, yet present complex overlapping patterns which may require sophisticated machine learning algorithms to decipher. Unfortunately, the limited availability of labeled spectra can hamper application of machine learning, especially deep learning algorithms which require large amounts of labeled data. In this context, simulation of spectral data becomes a tractable solution for algorithm development. RESULTS: Here, we introduce MetAssimulo 2.0, a comprehensive upgrade of the MetAssimulo 1.b metabolomic 1H NMR simulation tool, reimplemented as a Python-based web application. Where MetAssimulo 1.0 only simulated 1D 1H spectra of human urine, MetAssimulo 2.0 expands functionality to urine, blood, and cerebral spinal fluid, enhancing the realism of blood spectra by incorporating a broad protein background. This enhancement enables a closer approximation to real blood spectra, achieving a Pearson correlation of approximately 0.82. Moreover, this tool now includes simulation capabilities for 2D J-resolved (J-Res) and Correlation Spectroscopy spectra, significantly broadening its utility in complex mixture analysis. MetAssimulo 2.0 simulates both single, and groups, of spectra with both discrete (case-control, e.g. heart transplant versus healthy) and continuous (e.g. body mass index) outcomes and includes inter-metabolite correlations. It thus supports a range of experimental designs and demonstrating associations between metabolite profiles and biomedical responses.By enhancing NMR spectral simulations, MetAssimulo 2.0 is well positioned to support and enhance research at the intersection of deep learning and metabolomics. AVAILABILITY AND IMPLEMENTATION: The code and the detailed instruction/tutorial for MetAssimulo 2.0 is available at https://github.com/yanyan5420/MetAssimulo_2.git. The relevant NMR spectra for metabolites are deposited in MetaboLights with accession number MTBLS12081.

MOTIVATION: Random forests (RFs) can deal with a large number of variables, achieve reasonable prediction scores, and yield highly interpretable feature importance values. As such, RFs are appropriate models for feature selection and further dimension reduction. However, RFs are often not appropriate for correlated datasets due to their mode of selecting individual features for splitting. Addressing correlation relationships in high-dimensional datasets is imperative for reducing the number of variables that are assigned high importance, hence making the dimension reduction most efficient. Here, we propose the LAtent VAriable Stochastic Ensemble of Trees (LAVASET) method that derives latent variables based on the distance characteristics of each feature and aims to incorporate the correlation factor in the splitting step. RESULTS: Without compromising on performance in the majority of examples, LAVASET outperforms RF by accurately determining feature importance across all correlated variables and ensuring proper distribution of importance values. LAVASET yields mostly non-inferior prediction accuracies to traditional RFs when tested in simulated and real 1D datasets, as well as more complex and high-dimensional 3D datatypes. Unlike traditional RFs, LAVASET is unaffected by single 'important' noisy features (false positives), as it considers the local neighbourhood. LAVASET, therefore, highlights neighbourhoods of features, reflecting real signals that collectively impact the model's predictive ability. AVAILABILITY AND IMPLEMENTATION: LAVASET is freely available as a standalone package from https://github.com/melkasapi/LAVASET.

Metabolic profiling (metabolomics) aims at measuring small molecules (metabolites) in complex samples like blood or urine for human health studies. While biomarker-based assessment often relies on a single molecule, metabolic profiling combines several metabolites to create a more complex and more specific fingerprint of the disease. However, in contrast to genomics, there is no unique metabolomics setup able to measure the entire metabolome. This challenge leads to tedious and resource consuming preliminary studies to be able to design the right metabolomics experiment. In that context, computer assisted metabolic profiling can be of strong added value to design metabolomics studies more quickly and efficiently. We propose a constraint-based modelling approach which predicts in silico profiles of metabolites that are more likely to be differentially abundant under a given metabolic perturbation (e.g. due to a genetic disease), using flux simulation. In genome-scale metabolic networks, the fluxes of exchange reactions, also known as the flow of metabolites through their external transport reactions, can be simulated and compared between control and disease conditions in order to calculate changes in metabolite import and export. These import/export flux differences would be expected to induce changes in circulating biofluid levels of those metabolites, which can then be interpreted as potential biomarkers or metabolites of interest. In this study, we present SAMBA (SAMpling Biomarker Analysis), an approach which simulates fluxes in exchange reactions following a metabolic perturbation using random sampling, compares the simulated flux distributions between the baseline and modulated conditions, and ranks predicted differentially exchanged metabolites as potential biomarkers for the perturbation. We show that there is a good fit between simulated metabolic exchange profiles and experimental differential metabolites detected in plasma, such as patient data from the disease database OMIM, and metabolic trait-SNP associations found in mGWAS studies. These biomarker recommendations can provide insight into the underlying mechanism or metabolic pathway perturbation lying behind observed metabolite differential abundances, and suggest new metabolites as potential avenues for further experimental analyses.

As terabytes of multi-omics data are being generated, there is an ever-increasing need for methods facilitating the integration and interpretation of such data. Current multi-omics integration methods typically output lists, clusters, or subnetworks of molecules related to an outcome. Even with expert domain knowledge, discerning the biological processes involved is a time-consuming activity. Here we propose PathIntegrate, a method for integrating multi-omics datasets based on pathways, designed to exploit knowledge of biological systems and thus provide interpretable models for such studies. PathIntegrate employs single-sample pathway analysis to transform multi-omics datasets from the molecular to the pathway-level, and applies a predictive single-view or multi-view model to integrate the data. Model outputs include multi-omics pathways ranked by their contribution to the outcome prediction, the contribution of each omics layer, and the importance of each molecule in a pathway. Using semi-synthetic data we demonstrate the benefit of grouping molecules into pathways to detect signals in low signal-to-noise scenarios, as well as the ability of PathIntegrate to precisely identify important pathways at low effect sizes. Finally, using COPD and COVID-19 data we showcase how PathIntegrate enables convenient integration and interpretation of complex high-dimensional multi-omics datasets. PathIntegrate is available as an open-source Python package.

BACKGROUND: Single sample pathway analysis (ssPA) transforms molecular level omics data to the pathway level, enabling the discovery of patient-specific pathway signatures. Compared to conventional pathway analysis, ssPA overcomes the limitations by enabling multi-group comparisons, alongside facilitating numerous downstream analyses such as pathway-based machine learning. While in transcriptomics ssPA is a widely used technique, there is little literature evaluating its suitability for metabolomics. Here we provide a benchmark of established ssPA methods (ssGSEA, GSVA, SVD (PLAGE), and z-score) alongside the evaluation of two novel methods we propose: ssClustPA and kPCA, using semi-synthetic metabolomics data. We then demonstrate how ssPA can facilitate pathway-based interpretation of metabolomics data by performing a case-study on inflammatory bowel disease mass spectrometry data, using clustering to determine subtype-specific pathway signatures. RESULTS: While GSEA-based and z-score methods outperformed the others in terms of recall, clustering/dimensionality reduction-based methods provided higher precision at moderate-to-high effect sizes. A case study applying ssPA to inflammatory bowel disease data demonstrates how these methods yield a much richer depth of interpretation than conventional approaches, for example by clustering pathway scores to visualise a pathway-based patient subtype-specific correlation network. We also developed the sspa python package (freely available at https://pypi.org/project/sspa/ ), providing implementations of all the methods benchmarked in this study. CONCLUSION: This work underscores the value ssPA methods can add to metabolomic studies and provides a useful reference for those wishing to apply ssPA methods to metabolomics data.

BACKGROUND: The search for statistically significant relationships between molecular markers and outcomes is challenging when dealing with high-dimensional, noisy and collinear multivariate omics data, such as metabolomic profiles. Permutation procedures allow for the estimation of adjusted significance levels without assuming independence among metabolomic variables. Nevertheless, the complex non-normal structure of metabolic profiles and outcomes may bias the permutation results leading to overly conservative threshold estimates i.e. lower than those from a Bonferroni or Sidak correction. METHODS: Within a univariate permutation procedure we employ parametric simulation methods based on the multivariate (log-)Normal distribution to obtain adjusted significance levels which are consistent across different outcomes while effectively controlling the type I error rate. Next, we derive an alternative closed-form expression for the estimation of the number of non-redundant metabolic variates based on the spectral decomposition of their correlation matrix. The performance of the method is tested for different model parametrizations and across a wide range of correlation levels of the variates using synthetic and real data sets. RESULTS: Both the permutation-based formulation and the more practical closed form expression are found to give an effective indication of the number of independent metabolic effects exhibited by the system, while guaranteeing that the derived adjusted threshold is stable across outcome measures with diverse properties.

Over-representation analysis (ORA) is one of the commonest pathway analysis approaches used for the functional interpretation of metabolomics datasets. Despite the widespread use of ORA in metabolomics, the community lacks guidelines detailing its best-practice use. Many factors have a pronounced impact on the results, but to date their effects have received little systematic attention. Using five publicly available datasets, we demonstrated that changes in parameters such as the background set, differential metabolite selection methods, and pathway database used can result in profoundly different ORA results. The use of a non-assay-specific background set, for example, resulted in large numbers of false-positive pathways. Pathway database choice, evaluated using three of the most popular metabolic pathway databases (KEGG, Reactome, and BioCyc), led to vastly different results in both the number and function of significantly enriched pathways. Factors that are specific to metabolomics data, such as the reliability of compound identification and the chemical bias of different analytical platforms also impacted ORA results. Simulated metabolite misidentification rates as low as 4% resulted in both gain of false-positive pathways and loss of truly significant pathways across all datasets. Our results have several practical implications for ORA users, as well as those using alternative pathway analysis methods. We offer a set of recommendations for the use of ORA in metabolomics, alongside a set of minimal reporting guidelines, as a first step towards the standardisation of pathway analysis in metabolomics.

MOTIVATION: Liquid chromatography-mass spectrometry (LC-MS) is a standard method for proteomics and metabolomics analysis of biological samples. Unfortunately, it suffers from various changes in the retention times (RT) of the same compound in different samples, and these must be subsequently corrected (aligned) during data processing. Classic alignment methods such as in the popular XCMS package often assume a single time-warping function for each sample. Thus, the potentially varying RT drift for compounds with different masses in a sample is neglected in these methods. Moreover, the systematic change in RT drift across run order is often not considered by alignment algorithms. Therefore, these methods cannot effectively correct all misalignments. For a large-scale experiment involving many samples, the existence of misalignment becomes inevitable and concerning. RESULTS: Here, we describe an integrated reference-free profile alignment method, neighbor-wise compound-specific Graphical Time Warping (ncGTW), that can detect misaligned features and align profiles by leveraging expected RT drift structures and compound-specific warping functions. Specifically, ncGTW uses individualized warping functions for different compounds and assigns constraint edges on warping functions of neighboring samples. Validated with both realistic synthetic data and internal quality control samples, ncGTW applied to two large-scale metabolomics LC-MS datasets identifies many misaligned features and successfully realigns them. These features would otherwise be discarded or uncorrected using existing methods. The ncGTW software tool is developed currently as a plug-in to detect and realign misaligned features present in standard XCMS output. AVAILABILITY AND IMPLEMENTATION: An R package of ncGTW is freely available at Bioconductor and https://github.com/ChiungTingWu/ncGTW. A detailed user's manual and a vignette are provided within the package. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Metabolomics time-course experiments provide the opportunity to understand the changes to an organism by observing the evolution of metabolic profiles in response to internal or external stimuli. Along with other omic longitudinal profiling technologies, these techniques have great potential to uncover complex relations between variations across diverse omic variables and provide unique insights into the underlying biology of the system. However, many statistical methods currently used to analyse short time-series omic data are i) prone to overfitting, ii) do not fully take into account the experimental design or iii) do not make full use of the multivariate information intrinsic to the data or iv) are unable to uncover multiple associations between different omic data. The model we propose is an attempt to i) overcome overfitting by using a weakly informative Bayesian model, ii) capture experimental design conditions through a mixed-effects model, iii) model interdependencies between variables by augmenting the mixed-effects model with a conditional auto-regressive (CAR) component and iv) identify potential associations between heterogeneous omic variables by using a horseshoe prior. RESULTS: We assess the performance of our model on synthetic and real datasets and show that it can outperform comparable models for metabolomic longitudinal data analysis. In addition, our proposed method provides the analyst with new insights on the data as it is able to identify metabolic biomarkers related to treatment, infer perturbed pathways as a result of treatment and find significant associations with additional omic variables. We also show through simulation that our model is fairly robust against inaccuracies in metabolite assignments. On real data, we demonstrate that the number of profiled metabolites slightly affects the predictive ability of the model. CONCLUSIONS: Our single model approach to longitudinal analysis of metabolomics data provides an approach simultaneously for integrative analysis and biomarker discovery. In addition, it lends better interpretation by allowing analysis at the pathway level. An accompanying R package for the model has been developed using the probabilistic programming language Stan. The package offers user-friendly functions for simulating data, fitting the model, assessing model fit and postprocessing the results. The main aim of the R package is to offer freely accessible resources for integrative longitudinal analysis for metabolomics scientists and various visualization functions easy-to-use for applied researchers to interpret results.

BACKGROUND: Transcriptomic data is often used to build statistical models which are predictive of a given phenotype, such as disease status. Genes work together in pathways and it is widely thought that pathway representations will be more robust to noise in the gene expression levels. We aimed to test this hypothesis by constructing models based on either genes alone, or based on sample specific scores for each pathway, thus transforming the data to a 'pathway space'. We progressively degraded the raw data by addition of noise and examined the ability of the models to maintain predictivity. RESULTS: Models in the pathway space indeed had higher predictive robustness than models in the gene space. This result was independent of the workflow, parameters, classifier and data set used. Surprisingly, randomised pathway mappings produced models of similar accuracy and robustness to true mappings, suggesting that the success of pathway space models is not conferred by the specific definitions of the pathway. Instead, predictive models built on the true pathway mappings led to prediction rules with fewer influential pathways than those built on randomised pathways. The extent of this effect was used to differentiate pathway collections coming from a variety of widely used pathway databases. CONCLUSIONS: Prediction models based on pathway scores are more robust to degradation of gene expression information than the equivalent models based on ungrouped genes. While models based on true pathway scores are not more robust or accurate than those based on randomised pathways, true pathways produced simpler prediction rules, emphasizing a smaller number of pathways.

MOTIVATION: The field of toxicogenomics (the application of '-omics' technologies to risk assessment of compound toxicities) has expanded in the last decade, partly driven by new legislation, aimed at reducing animal testing in chemical risk assessment but mainly as a result of a paradigm change in toxicology towards the use and integration of genome wide data. Many research groups worldwide have generated large amounts of such toxicogenomics data. However, there is no centralized repository for archiving and making these data and associated tools for their analysis easily available. RESULTS: The Data Infrastructure for Chemical Safety Assessment (diXa) is a robust and sustainable infrastructure storing toxicogenomics data. A central data warehouse is connected to a portal with links to chemical information and molecular and phenotype data. diXa is publicly available through a user-friendly web interface. New data can be readily deposited into diXa using guidelines and templates available online. Analysis descriptions and tools for interrogating the data are available via the diXa portal. AVAILABILITY AND IMPLEMENTATION: http://www.dixa-fp7.eu CONTACT: [email protected]; [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Often during the analysis of biological data, it is of importance to interpret the correlation structure that exists between variables. Such correlations may reveal patterns of co-regulation that are indicative of biochemical pathways or common mechanisms of response to a related set of treatments. However, analyses of correlations are usually conducted by either subjective interpretation of the univariate covariance matrix or by applying multivariate modeling techniques, which do not take prior biological knowledge into account. Over-representation analysis (ORA) is a simple method for objectively deciding whether a set of variables of known or suspected biological relevance, such as a gene set or pathway, is more prevalent in a set of variables of interest than we expect by chance. However, ORA is usually applied to a set of variables differentiating a single experimental variable and does not take into account correlations. RESULTS: Over-representation of correlation analysis (ORCA) is a novel combination of ORA and correlation analysis that provides a means to test whether more associations exist between two specific groups of variables than expected by chance. The method is exemplified by application to drug sensitivity and microRNA expression data from a panel of cancer cell lines (NCI60). ORCA highlighted a previously reported correlation between sensitivity to alkylating anticancer agents and topoisomerase inhibitors. We also used this approach to validate microRNA clusters predicted by mRNA correlations. These observations suggest that ORCA has the potential to reveal novel insights from these data, which are not readily apparent using classical ORA. AVAILABILITY AND IMPLEMENTATION: The R code of the method is available at https://github.com/ORCABioinfo/ORCAcode.

MOTIVATION: Nuclear Magnetic Resonance (NMR) spectra are widely used in metabolomics to obtain metabolite profiles in complex biological mixtures. Common methods used to assign and estimate concentrations of metabolites involve either an expert manual peak fitting or extra pre-processing steps, such as peak alignment and binning. Peak fitting is very time consuming and is subject to human error. Conversely, alignment and binning can introduce artefacts and limit immediate biological interpretation of models. RESULTS: We present the Bayesian automated metabolite analyser for NMR spectra (BATMAN), an R package that deconvolutes peaks from one-dimensional NMR spectra, automatically assigns them to specific metabolites from a target list and obtains concentration estimates. The Bayesian model incorporates information on characteristic peak patterns of metabolites and is able to account for shifts in the position of peaks commonly seen in NMR spectra of biological samples. It applies a Markov chain Monte Carlo algorithm to sample from a joint posterior distribution of the model parameters and obtains concentration estimates with reduced error compared with conventional numerical integration and comparable to manual deconvolution by experienced spectroscopists. AVAILABILITY AND IMPLEMENTATION: http://www1.imperial.ac.uk/medicine/people/t.ebbels/ CONTACT: [email protected].

MOTIVATION: Metabolomics is the study of the complement of small molecule metabolites in cells, biofluids and tissues. Many metabolomic experiments are designed to compare changes observed over time under two experimental conditions or groups (e.g. a control and drug-treated group) with the goal of identifying discriminatory metabolites or biomarkers that characterize each condition. A common study design consists of repeated measurements taken on each experimental unit thus producing time courses of all metabolites. We describe a statistical framework for estimating time-varying metabolic profiles and their within-group variability and for detecting between-group differences. Specifically, we propose (i) a smoothing splines mixed effects (SME) model that treats each longitudinal measurement as a smooth function of time and (ii) an associated functional test statistic. Statistical significance is assessed by a non-parametric bootstrap procedure. RESULTS: The methodology has been extensively evaluated using simulated data and has been applied to real nuclear magnetic resonance spectroscopy data collected in a preclinical toxicology study as part of a larger project lead by the COMET (Consortium for Metabonomic Toxicology). Our findings are compatible with the previously published studies. AVAILABILITY: An R script is freely available for download at http://www2.imperial.ac.uk/~gmontana/sme.htm.

SUMMARY: Pathway-level analysis is a powerful approach enabling interpretation of post-genomic data at a higher level than that of individual biomolecules. Yet, it is currently hard to integrate more than one type of omics data in such an approach. Here, we present a web tool 'IMPaLA' for the joint pathway analysis of transcriptomics or proteomics and metabolomics data. It performs over-representation or enrichment analysis with user-specified lists of metabolites and genes using over 3000 pre-annotated pathways from 11 databases. As a result, pathways can be identified that may be disregulated on the transcriptional level, the metabolic level or both. Evidence of pathway disregulation is combined, allowing for the identification of additional pathways with changed activity that would not be highlighted when analysis is applied to any of the functional levels alone. The tool has been implemented both as an interactive website and as a web service to allow a programming interface. AVAILABILITY: The web interface of IMPaLA is available at http://impala.molgen.mpg.de. A web services programming interface is provided at http://impala.molgen.mpg.de/wsdoc. CONTACT: [email protected]; [email protected]; [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Using transcriptomic and metabolomic measurements from the NCI60 cell line panel, together with a novel approach to integration of molecular profile data, we show that the biochemical pathways associated with tumour cell chemosensitivity to platinum-based drugs are highly coincident, i.e. they describe a consensus phenotype. Direct integration of metabolome and transcriptome data at the point of pathway analysis improved the detection of consensus pathways by 76%, and revealed associations between platinum sensitivity and several metabolic pathways that were not visible from transcriptome analysis alone. These pathways included the TCA cycle and pyruvate metabolism, lipoprotein uptake and nucleotide synthesis by both salvage and de novo pathways. Extending the approach across a wide panel of chemotherapeutics, we confirmed the specificity of the metabolic pathway associations to platinum sensitivity. We conclude that metabolic phenotyping could play a role in predicting response to platinum chemotherapy and that consensus-phenotype integration of molecular profiling data is a powerful and versatile tool for both biomarker discovery and for exploring the complex relationships between biological pathways and drug response.

Abstract Motivation: High mass accuracy is an important goal in liquid chromatography–mass spectrometry experiments. Some manufacturers employ a mass calibration system that regularly switches between the analyte and a standard reference compound, and leads to gaps in the analyte data. We present a method for correction of such gaps in global molecular profiling applications such as metabolomics. We demonstrate that it improves peak detection and quantification, successfully recovering the expected number of peaks and intensity distribution in an example metabolomics dataset. Availability and implementation: Available in XCMS versions 1.23.3 and higher. Distributed via Bioconductor under GNU General Public License. (http://www.bioconductor.org/packages//2.7/bioc/html/xcms.html) Contact: [email protected]; [email protected]; [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: Probing the complex fusion of genetic and environmental interactions, metabolic profiling (or metabolomics/metabonomics), the study of small molecules involved in metabolic reactions, is a rapidly expanding 'omics' field. A major technique for capturing metabolite data is 1H-NMR spectroscopy and this yields highly complex profiles that require sophisticated statistical analysis methods. However, experimental data is difficult to control and expensive to obtain. Thus data simulation is a productive route to aid algorithm development. RESULTS: MetAssimulo is a MATLAB-based package that has been developed to simulate 1H-NMR spectra of complex mixtures such as metabolic profiles. Drawing data from a metabolite standard spectral database in conjunction with concentration information input by the user or constructed automatically from the Human Metabolome Database, MetAssimulo is able to create realistic metabolic profiles containing large numbers of metabolites with a range of user-defined properties. Current features include the simulation of two groups ('case' and 'control') specified by means and standard deviations of concentrations for each metabolite. The software enables addition of spectral noise with a realistic autocorrelation structure at user controllable levels. A crucial feature of the algorithm is its ability to simulate both intra- and inter-metabolite correlations, the analysis of which is fundamental to many techniques in the field. Further, MetAssimulo is able to simulate shifts in NMR peak positions that result from matrix effects such as pH differences which are often observed in metabolic NMR spectra and pose serious challenges for statistical algorithms. CONCLUSIONS: No other software is currently able to simulate NMR metabolic profiles with such complexity and flexibility. This paper describes the algorithm behind MetAssimulo and demonstrates how it can be used to simulate realistic NMR metabolic profiles with which to develop and test new data analysis techniques. MetAssimulo is freely available for academic use at http://cisbic.bioinformatics.ic.ac.uk/metassimulo/.

MOTIVATION: Metabolic profiles derived from high resolution (1)H-NMR data are complex, therefore statistical and machine learning approaches are vital for extracting useful information and biological insights. Focused modelling on targeted subsets of metabolites and samples can improve the predictive ability of models, and techniques such as genetic algorithms (GAs) have a proven utility in feature selection problems. The Consortium for Metabonomic Toxicology (COMET) obtained temporal NMR spectra of urine from rats treated with model toxins and stressors. Here, we develop a GA approach which simultaneously selects sets of samples and spectral regions from the COMET database to build robust, predictive classifiers of liver and kidney toxicity. RESULTS: The results indicate that using simultaneous sample and variable selection improved performance by over 9% compared with either method alone. Simultaneous selection also halved computation time. Successful classifiers repeatedly selected particular variables indicating that this approach can aid defining biomarkers of toxicity. Novel visualizations of the results from multiple computations were developed to aid the interpretability of which samples and variables were frequently selected. This method provides an efficient way to determine the most discriminatory variables and samples for any post-genomic dataset. AVAILABILITY: GA code available from http://www1.imperial.ac.uk/medicine/people/r.cavill/

BACKGROUND: Modelling the time-related behaviour of biological systems is essential for understanding their dynamic responses to perturbations. In metabolic profiling studies, the sampling rate and number of sampling points are often restricted due to experimental and biological constraints. RESULTS: A supervised multivariate modelling approach with the objective to model the time-related variation in the data for short and sparsely sampled time-series is described. A set of piecewise Orthogonal Projections to Latent Structures (OPLS) models are estimated, describing changes between successive time points. The individual OPLS models are linear, but the piecewise combination of several models accommodates modelling and prediction of changes which are non-linear with respect to the time course. We demonstrate the method on both simulated and metabolic profiling data, illustrating how time related changes are successfully modelled and predicted. CONCLUSION: The proposed method is effective for modelling and prediction of short and multivariate time series data. A key advantage of the method is model transparency, allowing easy interpretation of time-related variation in the data. The method provides a competitive complement to commonly applied multivariate methods such as OPLS and Principal Component Analysis (PCA) for modelling and analysis of short time-series data.