Cognitive analysis of metabolomics data for systems biology

AI/ML-driven advances in untargeted metabolomics and exposomics for biomedical applications

Metabolomics describes a high-throughput approach for measuring a repertoire of metabolites and small molecules in biological samples. One utility of untargeted metabolomics, unbiased global analysis of the metabolome, is to detect key metabolites as contributors to, or readouts of, human health and disease. In this perspective, we discuss how artificial intelligence (AI) and machine learning (ML) have promoted major advances in untargeted metabolomics workflows and facilitated pivotal findings in the areas of disease screening and diagnosis. We contextualize applications of AI and ML to the emerging field of high-resolution mass spectrometry (HRMS) exposomics, which unbiasedly detects endogenous metabolites and exogenous chemicals in human tissue to characterize exposure linked with disease outcomes. We discuss the state of the science and suggest potential opportunities for using AI and ML to improve data quality, rigor, detection, and chemical identification in untargeted metabolomics and exposomics studies.

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

Reviewing the metabolomics literature is becoming increasingly difficult because of the rapid expansion of relevant journal literature. Text-mining technologies are therefore needed to facilitate more efficient literature reviews. Here we contribute a standardised corpus of full-text publications from metabolomics studies and describe the development of two metabolite named entity recognition (NER) methods. These methods are based on Bidirectional Long Short-Term Memory (BiLSTM) networks and each incorporate different transfer learning techniques (for tokenisation and word embedding). Our first model (MetaboListem) follows prior methodology using GloVe word embeddings. Our second model exploits BERT and BioBERT for embedding and is named TABoLiSTM (Transformer-Affixed BiLSTM). The methods are trained on a novel corpus annotated using rule-based methods, and evaluated on manually annotated metabolomics articles. MetaboListem (F1-score 0.890, precision 0.892, recall 0.888) and TABoLiSTM (BioBERT version: F1-score 0.909, precision 0.926, recall 0.893) have achieved state-of-the-art performance on metabolite NER. A training corpus with full-text sentences from >1000 full-text Open Access metabolomics publications with 105,335 annotated metabolites was created, as well as a manually annotated test corpus (19,138 annotations). This work demonstrates that deep learning algorithms are capable of identifying metabolite names accurately and efficiently in text. The proposed corpus and NER algorithms can be used for metabolomics text-mining tasks such as information retrieval, document classification and literature-based discovery and are available from the omicsNLP GitHub repository.

Recent advances in metabolomics analysis for early drug development

The pharmaceutical industry adapted proteomics and other 'omics technologies for drug research early following their initial introduction. Although metabolomics lacked behind in this development, it has now become an accepted and widely applied approach in early drug development. Over the past few decades, metabolomics has evolved from a pure exploratory tool to a more mature and quantitative biochemical technology. Several metabolomics-based platforms are now applied during the early phases of drug discovery. Metabolomics analysis assists in the definition of the physiological response and target engagement (TE) markers as well as elucidation of the mode of action (MoA) of drug candidates under investigation. In this review, we highlight recent examples and novel developments of metabolomics analyses applied during early drug development.

A Checklist for Reproducible Computational Analysis in Clinical Metabolomics Research

Clinical metabolomics emerged as a novel approach for biomarker discovery with the translational potential to guide next-generation therapeutics and precision health interventions. However, reproducibility in clinical research employing metabolomics data is challenging. Checklists are a helpful tool for promoting reproducible research. Existing checklists that promote reproducible metabolomics research primarily focused on metadata and may not be sufficient to ensure reproducible metabolomics data processing. This paper provides a checklist including actions that need to be taken by researchers to make computational steps reproducible for clinical metabolomics studies. We developed an eight-item checklist that includes criteria related to reusable data sharing and reproducible computational workflow development. We also provided recommended tools and resources to complete each item, as well as a GitHub project template to guide the process. The checklist is concise and easy to follow. Studies that follow this checklist and use recommended resources may facilitate other researchers to reproduce metabolomics results easily and efficiently.

Metabolite discovery: Biochemistry's scientific driver

Metabolite identification represents a major challenge, and opportunity, for biochemistry. The collective characterization and quantification of metabolites in living organisms, with its many successes, represents a major biochemical knowledgebase and the foundation of metabolism's rebirth in the 21^st century; yet, characterizing newly observed metabolites has been an enduring obstacle. Crystallography and NMR spectroscopy have been of extraordinary importance, although their applicability in resolving metabolism's fine structure has been restricted by their intrinsic requirement of sufficient and sufficiently pure materials. Mass spectrometry has been a key technology, especially when coupled with high-performance separation technologies and emerging informatic and database solutions. Even more so, the collective of artificial intelligence technologies are rapidly evolving to help solve the metabolite characterization conundrum. This perspective describes this challenge, how it was historically addressed, and how metabolomics is evolving to address it today and in the future.

Integrated Metabolomics and Network Pharmacology to Establish the Action Mechanism of Qingrekasen Granule for Treating Nephrotic Syndrome

Nephrotic syndrome (NS) is a clinical syndrome resulting from abnormal glomerular permeability, mainly manifesting as edema and proteinuria. Qingrekasen granule (QRKSG), a Chinese Uyghur folk medicine, is a single-flavor preparation made from chicory (Cichorium intybus L.), widely used in treating dysuria and edema. Chicory, the main component in QRKSG, effectively treats edema and protects kidneys. However, the active components in QRKSG and its underlying mechanism for treating NS remain unclear. This study explored the specific mechanism and composition of QRKSG on an NS rat model using integrated metabolomics and network pharmacology. First, metabolomics explored the relevant metabolic pathways impacted by QRKSG in the treatment of NS. Secondly, network pharmacology further explored the possible metabolite targets. Afterward, a comprehensive network was constructed using the results from the network pharmacology and metabolomics analysis. Finally, the interactions between the active components and targets were predicted by molecular docking, and the differential expression levels of the target protein were verified by Western blotting. The metabolomics results showed “D-Glutamine and D-glutamate metabolism” and “Alanine, aspartate, and glutamate metabolism” as the main targeted metabolic pathways for treating NS in rats. AKT1, BCL2L1, CASP3, and MTOR were the core QRKSG targets in the treatment of NS. Molecular docking revealed that these core targets have a strong affinity for flavonoids, terpenoids, and phenolic acids. Moreover, the expression levels of p-PI3K, p-AKT1, p-mTOR, and CASP3 in the QRKSG group significantly decreased, while BCL2L1 increased compared to the model group. These findings established the underlying mechanism of QRKSG, such as promoting autophagy and anti-apoptosis through the expression of AKT1, CASP3, BCL2L1, and mTOR to protect podocytes and maintain renal tubular function.

Sample-to-analysis platform for rapid intracellular mass spectrometry from small numbers of cells

Real-time, advanced diagnostics of the biochemical state within cells remains a significant challenge for research and development, production, and application of cell-based therapies. The fundamental biochemical processes and mechanisms of action of such advanced therapies are still largely unknown, including the critical quality attributes that correlate to therapeutic function, performance, and potency and the critical process parameters that impact quality throughout cell therapy manufacturing. An integrated microfluidic platform has been developed for in-line analysis of a small number of cells via direct infusion nano-electrospray ionization mass spectrometry. Central to this platform is a microfabricated cell processing device that prepares cells from limited sample volumes removed directly from cell culture systems. The sample-to-analysis workflow overcomes the labor intensive, time-consuming, and destructive nature of existing mass spectrometry approaches for analysis of cells. By providing rapid, high-throughput analyses of the intracellular state, this platform enables untargeted discovery of critical quality attributes and their real-time, in-process monitoring.

Soil Salinity, a Serious Environmental Issue and Plant Responses: A Metabolomics Perspective

The effects of global warming have increasingly led to devastating environmental stresses, such as heat, salinity, and drought. Soil salinization is a serious environmental issue and results in detrimental abiotic stress, affecting 7% of land area and 33% of irrigated lands worldwide. The proportion of arable land facing salinity is expected to rise due to increasing climate change fuelled by anthropogenic activities, exacerbating the threat to global food security for the exponentially growing populace. As sessile organisms, plants have evolutionarily developed mechanisms that allow ad hoc responses to salinity stress. The orchestrated mechanisms include signalling cascades involving phytohormones, kinases, reactive oxygen species (ROS), and calcium regulatory networks. As a pillar in a systems biology approach, metabolomics allows for comprehensive interrogation of the biochemistry and a deconvolution of molecular mechanisms involved in plant responses to salinity. Thus, this review highlights soil salinization as a serious environmental issue and points to the negative impacts of salinity on plants. Furthermore, the review summarises mechanisms regulating salinity tolerance on molecular, cellular, and biochemical levels with a focus on metabolomics perspectives. This critical synthesis of current literature is an opportunity to revisit the current models regarding plant responses to salinity, with an invitation to further fundamental research for novel and actionable insights.

Deciphering Microbial Metal Toxicity Responses via Random Bar Code Transposon Site Sequencing and Activity-Based Metabolomics

To uncover metal toxicity targets and defense mechanisms of the facultative anaerobe Pantoea sp. strain MT58 (MT58), we used a multiomic strategy combining two global techniques, random bar code transposon site sequencing (RB-TnSeq) and activity-based metabolomics. MT58 is a metal-tolerant Oak Ridge Reservation (ORR) environmental isolate that was enriched in the presence of metals at concentrations measured in contaminated groundwater at an ORR nuclear waste site. The effects of three chemically different metals found at elevated concentrations in the ORR contaminated environment were investigated: the cation Al³⁺, the oxyanion CrO₄^2-, and the oxycation UO₂²⁺. Both global techniques were applied using all three metals under both aerobic and anaerobic conditions to elucidate metal interactions mediated through the activity of metabolites and key genes/proteins. These revealed that Al³⁺ binds intracellular arginine, CrO₄^2- enters the cell through sulfate transporters and oxidizes intracellular reduced thiols, and membrane-bound lipopolysaccharides protect the cell from UO₂²⁺ toxicity. In addition, the Tol outer membrane system contributed to the protection of cellular integrity from the toxic effects of all three metals. Likewise, we found evidence of regulation of lipid content in membranes under metal stress. Individually, RB-TnSeq and metabolomics are powerful tools to explore the impact various stresses have on biological systems. Here, we show that together they can be used synergistically to identify the molecular actors and mechanisms of these pertubations to an organism, furthering our understanding of how living systems interact with their environment. IMPORTANCE Studying microbial interactions with their environment can lead to a deeper understanding of biological molecular mechanisms. In this study, two global techniques, RB-TnSeq and activity metabolomics, were successfully used to probe the interactions between a metal-resistant microorganism, Pantoea sp. strain MT58, and metals contaminating a site where the organism can be located. A number of novel metal-microbe interactions were uncovered, including Al³⁺ toxicity targeting arginine synthesis, which could lead to a deeper understanding of the impact Al³⁺ contamination has on microbial communities as well as its impact on higher-level organisms, including plants for whom Al³⁺ contamination is an issue. Using multiomic approaches like the one described here is a way to further our understanding of microbial interactions and their impacts on the environment overall.

MSCAT: A Machine Learning Assisted Catalog of Metabolomics Software Tools

The bottleneck for taking full advantage of metabolomics data is often the availability, awareness, and usability of analysis tools. Software tools specifically designed for metabolomics data are being developed at an increasing rate, with hundreds of available tools already in the literature. Many of these tools are open-source and freely available but are very diverse with respect to language, data formats, and stages in the metabolomics pipeline. To help mitigate the challenges of meeting the increasing demand for guidance in choosing analytical tools and coordinating the adoption of best practices for reproducibility, we have designed and built the MSCAT (Metabolomics Software CATalog) database of metabolomics software tools that can be sustainably and continuously updated. This database provides a survey of the landscape of available tools and can assist researchers in their selection of data analysis workflows for metabolomics studies according to their specific needs. We used machine learning (ML) methodology for the purpose of semi-automating the identification of metabolomics software tool names within abstracts. MSCAT searches the literature to find new software tools by implementing a Named Entity Recognition (NER) model based on a neural network model at the sentence level composed of a character-level convolutional neural network (CNN) combined with a bidirectional long-short-term memory (LSTM) layer and a conditional random fields (CRF) layer. The list of potential new tools (and their associated publication) is then forwarded to the database maintainer for the curation of the database entry corresponding to the tool. The end-user interface allows for filtering of tools by multiple characteristics as well as plotting of the aggregate tool data to monitor the metabolomics software landscape.