EMN Webinars

From Metabolomics Society Wiki
Revision as of 08:16, 5 January 2021 by Viniciusveri (talk | contribs)
Jump to: navigation, search

The Early-Career Members Network (EMN) Committee has organized webinars since 2015 on a variety of metabolomics-related topics. This page provides a summary of all webinars to date with links to the webinar where available.

Contents

2015

Metabolomics: only suitable for multidisciplinary teams

By: Dr. Oscar Yanes

Date: 29 Jan 2015

Metabolomics is defined as the comprehensive and quantitative analysis of metabolites in living organisms. Among the omic sciences, metabolomics is possibly the most multidisciplinary of all, involving knowledge about electronic engineering and signal processing, analytical and organic chemistry, biostatistics and statistical physics, and biochemistry and cell metabolism. Here an untargeted metabolomics workflow will be detailed that provides examples of this multidisciplinarity.


Large-scale, computational and empirical UHPLC-MS-SPE-NMR annotation of plant metabolomes

By: Dr. Lloyd Sumner

Date: 5 March 2015

Integrated metabolomics is a revolutionary systems biology tool for understanding plant metabolism and elucidating gene function. Although the vast utility of metabolomics is well documented in the literature, its full scientific promise has not yet been realized due to multiple technical challenges. The number one, grand challenge of metabolomics is the large-scale confident chemical identification of metabolites. To address this challenge, we have developed sophisticated computational and empirical metabolomics tools for the systematic and biological directed annotation of plant metabolomes. This presentation will introduce novel software entitled Plant Metabolite Annotation Toolbox (PlantMAT) and a sophisticated UHPLC-MS-SPENMR instrumental ensemble that are being used for ‘sequencing’ the first plant metabolomes of the model plant systems Arabidopsis and Medicago truncatula.


Metabolomics and systems biology: models as guide and natural integrator

By: Prof. Bas Teusink

Date: 14 April 2015

In this webinar, I will discuss the use of mathematical models in guiding targeted and (semi-)untargeted metabolomics efforts. I will show how genome-scale metabolic models can be used as a data integration platform - also for metabolomics data. I will provide an example from medium optimisation in a biotechnological context. My message is that the use of models upfront combined with quantitative and well-timed metabolomics is often much more effective than simply generating lots of data and subsequent statistical analysis.


Metabolomics with high resolution mass spectrometry: from spectral databases to biomarker discovery

By: Dr. Christophe Junot

Date: 12 June 2015

Since the middle of the 2000s, high resolution mass spectrometry (HRMS) is widely used in metabolomics, mainly because the detection and identification of metabolites are improved compared to low resolution instruments. Thanks to their versatility, HRMS instruments are the most appropriate to achieve an optimal metabolome coverage, at the border of other omics fields such as lipidomics and glycomics. The aim of this talk will be to present HRMS based tools for metabolomics and lipidomics developed at the laboratory, and their relevance to the field of biomarker discovery for the diagnosis and follow-up of pathologies.


Metabolomics and Beyond: Challenges and Strategies for Next-gen Omic Analyses

By: Dr. Dmitry Grapov

Date: September 15, 2015

Robust interpretation of experimental results measuring discreet biological domains remains a significant challenge in the face of complex biochemical regulation processes such as organismal versus tissue versus cellular metabolism, epigenetics, and protein post-translational modification. Integration of analyses carried out across multiple measurement or omic platforms is an emerging approach to help address these challenges. The following presentation will focus on key challenges faced by metabolomics researchers in the areas large-scale studies data normalization, multivariate analysis, visualization and omics data integration.


2016

How close are we to metabolomics reproducibility? Data sharing, data standards and workflows for metabolomics

By: Dr. Reza Salek

Date: March 24, 2016


Chromatography and Metabolomics: What’s the Point?

By: Dr. Karl Burgess

Date: April 29, 2016

The key to both chromatography and mass spectrometry is the separation of chemical species based on their physicochemical properties. As metabolomics researchers, we look to improved chromatography so enable us to detect compounds we previously had trouble with, to reduce the enormous complexity of the samples we analyse, and to clean up samples before or during analysis. Advances in mass spectrometry bring us greater sensitivity, better selectivity and a toolbox of techniques to aid in identification of biochemicals. With all these advantages come many disadvantages - poor reproducibility, compound bias and contamination. In this webinar, I'll explore chromatography and mass spectrometry with a critical eye. How can we improve them? Do we need them in every experiment? And really, what's the point of metabolomics at all?


From Features to Compounds: How To Get Started and Move Forward

By: Dr. Jan Stanstrup

Date: May 27, 2016

In the untargeted analysis of complex mixtures the identification of compounds is the fundamental step enabling the results to be put in biological context. It is therefore of crucial importance for early career scientists approaching the field of metabolomics to familiarize themselves with this step. Many tools have been developed to aid identification; however, compound identification still constitutes one of the main bottlenecks in metabolomics and still requires substantial amounts of manual work. This webinar will go through the basic concepts used in MS-based compound identification and will introduce a number of relevant tools and databases allowing researcher to approach identification in a systematic way. The webinar is in particular dedicated to those researchers coming into the field of metabolomics that are not familiar with the methods used for compound identification and for which getting started can be a daunting and time-consuming challenge.


Targeted High-throughput Lipidomics Platform: Development and Application

By: Dr. Peter Meikle

Date: July 27, 2016

Lipidomics, full analysis of lipid species and their biological roles with respect to health and diseases, has attracted increasing attention of biological and analytical scientific community. Knowing and understanding steps involved in lipidomics experimental workflow is essential for successful outcome. The Metabolomics Laboratory at Baker IDI Heart and Diabetes Institute (Melbourne, Australia) has a focus on the dyslipidemia and altered lipid metabolism associated with obesity, diabetes and cardiovascular disease and its relationship to the pathogenesis of these disease states. The laboratory has developed a targeted lipidomics platform that is able to quantify over 500 lipid species in 15 minutes using liquid chromatography coupled to electrospray ionisation tandem mass spectrometry. This platform is being applied to identify new approaches to early diagnosis and risk assessment as well as the development of new lipid modulating therapies for chronic disease. With illustration from this well-established lipidomics platform, this webinar will discuss the development of targeted high-throughput lipidomics platform, including selection and characterisation of lipid species, development of chromatography and quality control in the analysis of large sample sets. The presentation will draw on specific examples to highlight the application to large cohort studies and the Institute’s work to develop new therapeutic strategies for cardiovascular disease.


Mass Spectral Libraries

By: Dr. Emma Schymanski

Date: October 6, 2016

Mass spectrometry is applied in diverse ways in metabolomics research and the mass spectrum of a small molecule can act as a fingerprint for identification. Just as the scientific questions in metabolomics vary, there is a diverse set of mass spectral libraries available to assist in the identification of metabolites and other small molecules. This webinar aims to provide listeners with a brief overview of several different mass spectral resources, including a personal view on pros and cons of the different options – providing a basis for listeners to choose the resource(s) that may best suit their investigation and needs. Additional information about substance overlap, spectral matching, identification confidence and spectral exchange will also be given – as well as some factors to consider carefully. Finally, some perspectives towards in silico identification without spectral libraries will be given to lead into a topic for a future webinar.


Smarter ways to clean LC-MS

By: Assoc. Prof. Carl Brunius

Date: November 17, 2016

LC-MS is the most frequently used technique for untargeted metabolomics owing to broad coverage of metabolites, high sensitivity and simple sample preparation. However, instrument data tends to have high noise contribution from drift in signal intensity, mass accuracy and retention times. This noise has both within batch and between batch contributions and results in reduced measurement repeatability and reproducibility. The power to detect biological responses may thus be decreased and interpretations consequently obscured. Dr. Carl Brunius (the speaker) is involved in developing procedures to address and correct for within- and between-batch variability in processing multiple-batch untargeted LC-MS metabolomics data to increase their quality. This webinar session will talk about (i) alignment and merging of LC-MS features that are systematically misaligned between batches and (ii) within batch intensity drift correction that allows multiple drift patterns within batch. These algorithms will be applied on authentic data, resulting in improved peak picking performance and decreased noise in the dataset. All algorithms are developed as open source and are, together with example data, freely available as an R package from https://GitLab.com/CarlBrunius/batchCorr.

2017

How well do I quantify? Concepts for method validation and evaluation of measurement uncertainty in metabolomics

By: Assoc. Prof. Stephan Hann

Date: March 24, 2017

This webinar will give an introduction to the basic terminology and principles of validation and measurement uncertainty in metabolomics. It will be demonstrated how validation parameters are determined in selected examples (e.g. LC-MS/MS, GC-MS/MS) for quantitative metabolomic analysis. Different quantification approaches will be overviewed in detail, and tips on choosing the most appropriate analytical strategies to answer metabolome-level hypotheses will be given. Finally, a workflow for estimating measurement uncertainty in metabolomics will be presented and discussed.


Machine learning powered metabolomic network analysis

By: Dr. Dmitry Grapov

Date: May 30, 2017

Metabolomic network analysis can be used to interpret experimental results within a variety of contexts including: biochemical relationships, structural and spectral similarity and empirical correlation. Machine learning is useful for modeling relationships in the context of pattern recognition, clustering, classification and regression based predictive modeling. The combination of developed metabolomic networks and machine learning based predictive models offer a unique method to visualize empirical relationships while testing key experimental hypotheses. The following presentation focuses on data analysis, visualization, machine learning and network mapping approaches used to create richly mapped metabolomic networks.


Using PhenoMeNal for your metabolomics data analysis on the cloud

By: Dr. Pablo Moreno

Date: October 24, 2017

PhenoMeNal aims to bridge the gap between cloud computing and metabolomics researchers by providing the ability to create Cloud Research Environments (CRE) for metabolomics data analysis. A PhenoMeNal CRE is a small cluster of computers with popular metabolomics data analysis tools already installed. These tools are ready to be run and are accessible through a user friendly Galaxy workflow environment reducing the need for in-house bioinformatics. The PhenoMeNal CRE not only includes data analysis tools, but also example workflows where some of these tools are used together. You can also make your own workflows inside the CRE. In this webinar we will explain the main components of PhenoMeNal. We will demonstrate how to register, access existing tools and workflows, create a new PhenoMeNal CRE on Amazon, and execute a workflow on a PhenoMeNal CRE.


Introduction to Stable Isotope Resolved Metabolomics (SIRM) and Applications to Cancer Metabolism

By: Dr. Andrew Lane

Date: December 14, 2017

Stable isotope resolved metabolomics (SIRM), for pathway tracing, represents an important new approach to obtaining metabolic parameters. SIRM allows the generation of atomic fate maps in cells and tissues, which provides the necessary information and data for metabolic flux analyis. This powerful new approach has already provided important new insights into metabolic adaptations in lung cancer cells.


2018

Metabolomics as a hypothesis generator for understanding metabolic regulation

By: Prof. Uwe Sauer

Date: February 14, 2018

Prof. Uwe Sauer focuses on two conceptual problems: i) discovery of functionally important regulation mechanisms and ii) understanding which of the many known mechnisns actually matter for a given adaption. On the discovery side, he illustrates the use of coarse-grained kinetic models to extract mechnistic hypotheses from dynamic metabolomics data. For learning the coordination mechanisms, he presents an approach that hypothesizes the dynamically important mechnism from the much fewer steady state measurements in the bacterium E. coli. The surprising result is that only very few regulation events appear to be required for a given transition, typically involving less than a handful of active regulators.


Can we capture an accurate view of cell or tissue specific metabolism from an expression profile?

By: Dr. Nathan Lewis

Date: March 26, 2018


Use better software: MS-DIAL instead of XCMS, ChemRich/MetDA instead of Metaboanalyst

By: Dr. Oliver Fiehn

Date: April 24, 2018

XCMS and MetaboAnalystare the two most popular tools used by metabolomics researchers. But are these the best options? The West Coast Metabolomics Center at UC Davis has collaborated with RIKEN/NGI in Japan to release MS-DIAL vs.2 that yields far fewer false positive peak detections in untargeted LC-MS/MS runs than XCMS, with superior integration of compound identification software. MS-DIAL now also works on low or high resolution GC-MS data, making it the tool of choice for raw data processing of any mass spectrometry-based metabolomics study.We have also developed alternative software suites for statistical analysis of final result data. ChemRich uses all identified metabolites, including complex lipids, for set enrichment statistics. In comparison, MetaboAnalyst cannot perform pathway enrichment statistics on more than half of all identified metabolites, because it relies on KEGG pathways only. Moreover, we have developed new software for improved data normalization and statistics workflows in MetDA web-based analyses that will be presented on published example data.


Activity Metabolomics: Technologies to identify metabolites that modulate phenotype

By: Prof. Gary Siuzdak

Date: May 29, 2018

Metabolomics is broadly acknowledged to be the omics discipline closest tothe phenotype and therefore widely used for biomarker discovery. However,metabolomics can also be designed to identify active metabolites thatalter a cell’s or an organism's phenotype.This "Activity Metabolomics”concept integrates metabolomics data analysis with pathway and systemsbiology data, ultimately to select endogenous metabolites that can bescreened for functionality. A growing literature reports the use ofmetabolites to modulate diverse processes including stem celldifferentiation, oligodendrocytematuration, insulin signaling,T-cellsurvival and macrophage immune responses. We have developed XCMS Online(xcmsonline.scripps.edu) and the newly expanded METLIN database (now withover 50,000 standards containing MS/MS data) to perform untargeted andtargeted metabolomics, as well as pathway analysis and systems biologydata integration.Metabolomics Activity Screening (MAS) has beenimplemented within XCMS Online to help achieve this integration goal foridentifying active metabolites. Because metabolites are often readilyavailable, activity metabolomics is uniquely positioning its practitionersto move beyond biomarkers, and become active participants in thebiological endgame of modulating phenotype. (for more information seeNature Biotechnology 2018 nature.com/articles/nbt.4101)


Metabolomics-based Elucidation of Plant Specialized Metabolism

By: Prof. Kazuki Saito

Date: July 25, 2018

The recent advances of genomics and metabolomics in plant science accelerate our understanding about the mechanism, regulation and evolution of biosynthesis of plant specialized products. We can now address the questions how the metabolomic diversity of plants is originated at the levels of genome (phytochemical genomics) and how we should apply this knowledge to drug discovery, industry and agriculture. In this presentation, at first, technological developments of metabolomic analysis will be discussed forthe better understanding chemical diversity of plants. Then, a couple of examples of application of metabolomics to functional genomics of specialized metabolism in a model plant Arabidopsis thalianawill be presented, focusing on the biosynthesis of phenylpropanoids and lipids. The further extension to crops and medicinal plants producing a variety of specialized metabolites will be presented.


Recent advances towards integrating metabolomics in kinetic models of large metabolic networks

By: Prof. Lars Nielsen

Date: October 15, 2018

Lars Nielsen is Chair of Biological Engineering at the Australian Institute for Bioengineering &Nanotechnology and Scientific Director for the Section for Quantitative Modelling of Cellular Metabolism at the Novo Nordisk Foundation Center for Biosustainability in Denmark. He is Director of the Bioplatforms Australia Queensland Node for Metabolomics and Proteomics, which provides systems and synthetic biology support to design and build cell factories for the production of fuels, chemicals and pharmaceuticals. His core research interest is modelling of cellular metabolism and his team has made many contributions to the formulation and use of genome scale models. He recently received a Novo Nordisk Foundation Laureate Research Grant to develop large scale, mathematical models to explore and explain the molecular basis for homeostasis–the self-regulating processes evolved to maintain metabolic equilibrium. Studying homeostasis is relevant for the understanding and treatment of complex diseases, particular with the emergence of personalized medicine. It is equally important when we seek to repurpose the cellular machinery for the production of desired chemicals, materials and pharmaceuticals.


2019

Untargeted metabolomics reveals smokers' characteristic profiles

By: Dr. Ping-Ching Hsu

Date: March 1, 2019


Multidimensional Chromatography, what is the synergy with untargeted metabolomics profiling?

By: Dr. Pierre-Hugues Stefanuto

Date: March 28, 2019

In this webinar, Dr Pierre-Hugues Stefanuto will discuss the new development of multidimensional chromatography and the synergy with metabolomics. This presentation will be broadcasted in the context of the Multidimensional Chromatography Workshop held in Liège last January (http://multidimensionalchromatography.com). During this event, four focus group discussions were organized: 1) data processing for untargeted screening, 2) minimum reporting information for QC and compound validation, 3) hyphenation of MDGC with high-resolution MS, 4) and the general acceptance of MDC techniques. He will illustrate these topics through some ongoing medical research articulated around volatile organic compound (VOC) measurements in human breath and in vitro in metabolomic applications.


Computational mass spectrometry in metabolomics to deepen the understanding of metabolisms

By: Dr. Hiroshi Tsugawa

Date: April 23, 2019

Computational mass spectrometry is a growing research field to process mass spectrometry data, assist the interpretation of mass fragmentations, and elucidate unknown structures with metabolome databases and repositories for the global identification of metabolomes in various living organisms. In this talk, Dr Tsugawa will introduce three metabolomics software programs which include (1) MS-DIAL for untargeted metabolomics, (2) MS-FINDER for structure elucidations of unknowns, and (3) MRMPROBS for targeted metabolomics. These programs are demonstrated to perform the comprehensive analyses of primary metabolites, lipids, and plant specialized metabolites where unknown metabolites are also untangled with various methodologies including stable isotope labeled organisms, metabolite class recommendations, and integrated metabolome network analyses. In addition, a computational workflow to link untargeted- and targeted metabolomics is also highlighted in this talk.


A Combined NMR and MS Metabolomics Approach to Study Neurodegenerative Diseases

By: Dr. Robert Powers

Date: May 24, 2019

The metabolome captures how the system responds to drug treatment, disease state, or genetic modification. In this regards, metabolomics is an invaluable approach to easily and rapidly diagnose human disease and to assist in personalized medicine by monitoring a patient’s response to treatments. But, metabolomics is deceivingly complex with numerous sources of errors and technical challenges at every step of the process. One specific challenge is achieving a complete and accurate coverage of the metabolome, which can be addressed by combining NMR and mass spectrometry. Our metabolomics protocols and MVAPACK software for integrating NMR and mass spectral data for the analysis of neurodegenerative disease will be discussed. Our investigation into the molecular mechanisms of Parkinson’s disease and the identification of biomarkers for multiple sclerosis will be highlighted.


Metabolomics 2019: Professional Career Development: The Survival Kit

By: Cathy Delhanty

Date: June 23, 2019


Lipidomics and epilipidomics – from analysis to data integration

By: Dr. Maria Fedorova Date: July 17, 2019

Lipidomics is a large-scale study of diversified molecular species of lipids aiming to address the identity, quantities, cellular and tissue distribution of lipids as well as related signalling and metabolic pathways. In addition to the native lipidome, lipid species derived from enzymatic and non-enzymatic modifications (the epilipidome) further increase regulatory capacity of the biological systems. High diversity of physico-chemical properties as well as large dynamic range of lipid concentrations in native lipidomes makes significantly challenge their analysis. For omics-wide high-throughput identification of lipid species from complex biological samples, several crucial analytical steps including extraction, chromatographic separation and mass spectrometry analysis need to be carefully considered and validated. The webinar will review current analytical strategies used in contemporary lipidomics and epilipidomics with the focus on optimization of LC-MS/MS based workflows for “discovery” lipidomics including sample preparation, lipid fractionation, separation using different chromatography techniques and high-throughput identification solutions. Available bioinformatics tools for identification of native and modified lipids will be described and compared as well as possible lipidomics data integration strategies.


Discovering Metabolites that Alter Physiology, an Omics Perspective

By: Dr. Gary Siuzdak Date: September 18, 2019

Metabolomics and the comprehensive analysis of the metabolome and lipidome has traditionally been pursued with the aim of identifying biomarkers in the diagnosis and prediction of disease. However, the value of metabolomics has been redefined from a simple biomarker identification tool to a technology for the discovery of active drivers of biological processes. It is now clear that the metabolome affects cellular physiology through modulation of other ‘omics’ levels, including the genome, epigenome, transcriptome and proteome. In this presentation, I will focus on our recent progress in using metabolomics to understand how the metabolome influences other omics and, by extension, to reveal the active role of metabolites in physiology and disease. This concept of utilizing metabolomics to perform activity screens to identify biologically active metabolites — which we term activity metabolomics — is already having a broad impact on biology.


Metabolomics as a tool for elucidating plant growth regulation

By: Dr. Camila Caldana Date: November 20, 2019

Rising demand for food and fuels makes it crucial to develop breeding strategies for increasing crop yield/biomass. Plant biomass production is tightly associated with growth and relies on a tight regulation of a complex signaling network that integrates external and internal stimuli. The main goal of our group is to elucidate the processes underlying plant growth and production of biomass by combining physiology, metabolomics, and gene expression analyses. In my presentation, I will provide examples of i) how the evolutionary conserved Target of Rapamycin pathway fine-tunes metabolic homeostasis to promote biosynthetic growth in plants; ii) the potential of metabolite profiles to predict plant performance as biomarkers.

2020

Human metabolic individuality and its role in multi-omics based studies

By: Dr. Karsten Suhre

Date: February 4, 2020

In this webinar, Dr. Suhre will discuss Genome-wide association studies with clinically relevant intermediate traits, such as gene expression, proteomics, and metabolomics, which unravelled numerous pathophysiological pathways and generated many hypotheses regarding the functional basis of complex disorders. More recently, similar approaches linked variation in epigenetic modifications, especially differential methylation of chromosomal CpG-pairs, to changes in gene expression and blood circulating metabolites. These large-scale population and patient cohort studies reflect experimental data obtained from naturally occurring variance of the general population where each individual may be viewed as an experiment conducted by Nature. The next and most challenging step on the way to a truly personalized approach to medicine is to translate the results from these large-scale omics studies to applications at the patient level. In this presentation, Dr. Suhre will present a general overview of where the field presently stands.


Plant Metabolomics - Elucidating biochemicals mechanisms underlying plant-environment interactions

By: Dr. Fidele Tugizimana

Date: May 1, 2020

In this webinar, Dr. Fidele will provide a snapshot of applications of metabolomics in plant sciences, particularly in plant-environment interactions research. The webinar will highlight some examples of the use of metabolomics to elucidate hypothetical frameworks that describe the biochemistry underlying naïve and primed-plant responses to microbial infections. Furthermore, one of the novel (emerging) strategies for sustainable food production and food security is the use of biostimulants in agriculture industry. Application of metabolomics in decoding and understanding plant-biostimulant interactions will be also highlighted.


Decomposing Complex Metabolite Mixtures...& Unraveling the neonatal metabolome

Date: June 19, 2020

Decomposing Complex Metabolite Mixtures through Substructure-based Metabolomics Workflows

By: Justin J.J. van der Hooft

In this webinar, Dr. Justin J.J. van der Hooft will start with an introduction on the challenges of metabolite annotation and identification in untargeted metabolomics experiments of complex mixtures typically encountered in natural products and food research. He will also show how MS2LDA has been successfully used to capture chemical knowledge from diverse plant, food, and microbial-related data sets. Dr. Justin J.J. van der Hooft will finish by highlighting the advantages of combining metabolome mining and annotation tools in public data of different plant and microbial-related studies.


Unraveling the neonatal metabolome using mass spectral data mining tools

By: Madeleine Ernst

In this webinar, Dr. Madeleine Ernst will explain how mass spectral data mining tools, such as molecular networking through the community platform GNPS, MS2LDA, in silico structure prediction (e.g. Network Annotation Propagation) and ClassyFire can significantly enhance chemical structural annotation retrieved in clinical mass spectrometry-based metabolomics studies. Dr. Madeleine Ernst will also elucidate how metabolic signatures of neonatal health and disease can be unraveled, significantly enhancing biological interpretation and hypothesis generation in metabolomics studies.

Metabolic networks

Date: November 26, 2020

Metabolic networks to enrich and interpret metabolic fingerprints

By: Dr. Fabien Jourdan

Metabolic modulation is a cornerstone cellular response to genetic or environmental stresses. This plasticity is going beyond central metabolism and may involve complex processes spanning several metabolic pathways. Hence, it is a key challenge to be able to decipher metabolic modulations in a systemic and global perspective. The aim of the computational methods and tools which will be presented is thus to consider the full complexity of metabolism. To do so, all metabolic reactions the cell is able to achieve are gathered in a single mathematical model call “genome scale metabolic network”. Based on this model, it is then possible to identify metabolic modulations associated to metabolic fingerprints or suggest metabolites of interest to enrich biochemical interpretation.


Improving metabolic studies with diverse context-specific metabolic networks

By: Dr. Pablo Rodriguez Mier

Understanding deregulations of metabolism based on isolated measures of gene expression, protein or metabolite concentrations is a challenging task due to the interconnection of multiple processes. One solution is to extract, from generic genome-scale metabolic networks, the specific sub-network which is modulated in the studied condition. Many algorithms have been proposed for such context-specific network extraction based on experimental measurements. However, this process is subject to some randomness and variability, since multiple metabolic networks can model the metabolic state in a similarly adequate manner for the same experimental data. This means that for some condition and reconstruction method, there are usually multiple possible sub-networks that can explain the same experimental data, but most current methods ignore this fact and return a single sub-network instead. Ignoring this variability can not only lead to incorrect or incomplete explanations of the biological experiment, but also causes valuable information to be lost that could be used to improve predictions. In this talk we will see what context-specific metabolic sub-networks are, some of the limitations of the current methods, and how we can get a diverse set of sub-networks to improve predictions and obtain better mechanistic insights about the metabolism.


New bio-statistical methods for metabolomics and metabolite annotation

Date: December 22, 2020

New bio-statistical methods for metabolomics

By: Dr. Daniel Raftery

Highly complex biological samples present challenging analysis problems for the field of metabolomics. Ideally, platforms that provide broad metabolome coverage and high data quality allow the opportunity for deep insights into biological problems. However, this goal can be difficult to achieve on a routine basis because the highly complex data are subject to matrix effects and complicated correlative relationships among many metabolites. As such metabolite identification, biomarker identification and validation can be very challenging. Advanced statistical methods are needed to deal with these issues for improved biomarker discovery, unknown identification and biological interpretation. We have pursued the development of a number of approaches that try to unravel the complex and multidimensional structure of metabolomics datasets, with some successes and some failures along the way. In this talk, I will provide some examples of where even non-experts in biostatistics can make progress in developing advanced analysis approaches and discuss some areas that provide significant challenges for future work.


Improving metabolic studies with diverse context-specific metabolic networks

By: Dr. Fausto Carnevale Neto

The major goal of metabolomics is to interrogate complex biological extracts for the purposes of metabolic exploration and biomarker discovery. Liquid chromatography (LC) coupled with tandem mass spectrometry (MS/MS or MS2) is widely accepted as a powerful strategy to explore the chemical constituency of complex mixtures. It offers accurate masses of detected ions and the structural information derived from fragmentation reactions in the gas-phase. While global LC-MS/MS profiling provides the comprehensive measurement of metabolites in complex biological samples, structure annotation remains a challenge, and computational approaches are necessary to translate the molecular composition into biological knowledge. Recently, global natural products social molecular networking (GNPS) has emerged as superb mining tool to assist the interpretation of large MS/MS datasets in the context of metabolomics. It integrates spectral database matching, unsupervised molecular substructure discovery, in silico fragmentation prediction, and automated chemical classification into a network topology. By embedding independent experimental and predictive annotation outputs on to the multi-informative molecular network, we can expand the automated chemical structural annotation within complex metabolic mixtures.