Search

Biomarker discovery, i.e., finding disease or condition-specific biological markers, is a crucial aspect of biomedical research. Volatile organic compounds (VOCs) are excreted by various biofluids, cells and tissues, and bacteria, and these have been investigated extensively for their potential as markers of malfunctioning status in human. The number of VOCs excreted by those media - typically detected using sophisticated analytical instrumentation - are numerically large and biologically complex. Therefore, data preprocessing and analysis are crucial for successful identification of valid VOC markers for their application in clinical practice. This chapter provides an overview of various preprocessing approaches suitable for volatilome data of diverse nature. The importance of normalization and scaling, often neglected in the field, is discussed. The most common and promising machine learning techniques are presented and discussed, including unsupervised and supervised approaches, followed by a rarely used strategy in the volatilomics field, data fusion. The chapter aims to equip the reader with a basic overview of suitable techniques for treating and successfully exploiting volatilome data.

It has been known for some time that vapours and odours produced by the body and breath can have some diagnostic value. This knowledge has been confirmed in recent history within clinical trial studies where animals have been successfully trained to detect diseases, through the sniffing of precise volatile organic profiles. Such metabolites can also be captured using analytical technologies. The term 'breathomics' is used to denote the science of analysing all the metabolites that are found in an organism’s breath. Studies in non-targeted metabolites with breathomics make use of Gas Chromatography-Mass Spectrometry (GC-MS) as the most advisable analytical method. Generally, the metabolic profiling of breath analysis encompasses the handling, configuration, scaling and bundling of thou-sands of features obtained from the GC-MS data obtained from hundreds of participating individuals. Also, as is the case with other technologies used in diagnosis,there are many random and systematic noises that influence breathomics. It is for this reason that multi-step data processing (deconvoluted) is needed. However,this is a process that is not only complex; it is also time-consuming and prone to operator errors. [Continues]

Metabolic profiling of breath analysis involves processing, alignment, scaling, and clustering of thousands of features extracted from gas chromatography/mass spectrometry (GC/MS) data from hundreds of participants. The multistep data processing is complicated, operator error-prone, and time-consuming. Automated algorithmic clustering methods that are able to cluster features in a fast and reliable way are necessary. These accelerate metabolic profiling and discovery platforms for next-generation medical diagnostic tools. Our unsupervised clustering technique, VOCCluster, prototyped in Python, handles features of deconvolved GC/MS breath data. VOCCluster was created from a heuristic ontology based on the observation of experts undertaking data processing with a suite of software packages. VOCCluster identifies and clusters groups of volatile organic compounds (VOCs) from deconvolved GC/MS breath with similar mass spectra and retention index profiles. VOCCluster was used to cluster more than 15 000 features extracted from 74 GC/MS clinical breath samples obtained from participants with cancer before and after a radiation therapy. Results were evaluated against a panel of ground truth compounds and compared to other clustering methods (DBSCAN and OPTICS) that were used in previous metabolomics studies. VOCCluster was able to cluster those features into 1081 groups (including endogenous and exogenous compounds and instrumental artifacts) with an accuracy rate of 96% (±0.04 at 95% confidence interval).

Through their breath, humans exhale hundreds of volatile organic compounds (VOCs) that can reveal pathologies, including many types of cancer at early stages. Gas chromatography-mass spectrometry (GC-MS) is an analytical method used to separate and detect compounds in the mixture contained in breath samples. The identification of VOCs is based on the recognition of their specific ion patterns in GC-MS data, which requires labour-intensive and time-consuming preprocessing and analysis by domain experts. This paper explores the original idea of applying supervised machine learning, and in particular convolutional neural networks (CNNs), to learn ion patterns directly from raw GC-MS data. The method adapts to machine specific characteristics, and once trained, can quickly analyse breath samples bypassing the time-consuming preprocessing phase. The CNN classification performance is compared to those of shallow neural networks and support vector machines. All considered machine learning tools achieved high accuracy in experiments with clinical data from participants. In particular, the CNN-based approach detected the lowest number of false positives. The results indicate that the proposed method is a promising tool to improve accuracy, specificity, and in particular speed in the detection of VOCs of interest in large-scale data analysis.