Metabolomics in the Age of Cloud Computing, AI and Machine Learning

A special issue of Metabolites (ISSN 2218-1989). This special issue belongs to the section "Bioinformatics and Data Analysis".

Deadline for manuscript submissions: closed (30 June 2022) | Viewed by 4987

Special Issue Editors

BiotechVision, Cambridge CB22 3BG, UK
Interests: bioinformatics; computational biology; systems biology; metabolomics
Special Issues, Collections and Topics in MDPI journals
Bioinformatics Group, Department of Plant Sciences, Wageningen University, 6708 PB Wageningen, The Netherlands
Interests: metabolomics; metabolite annotation; metabolite identification; metabolome mining; mass spectrometry; mass fragmentation; machine learning-based approaches; substructures; chemical classes; natural product discovery; food metabolome
Special Issues, Collections and Topics in MDPI journals
Department of Computer Science, Tufts University, Medford and Somerville, MA, USA
Interests: metabolomics; systems biology; machine learning
School of Computing Science, University of Glasgow, Glasgow, UK
Interests: machine learning; metabolomics; mass spectrometry data acquisition; mass spectrometry data analysis; computational biology
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

There have been several recent advancements in computer science, computational biology, and machine learning for large-scale analysis and interpretation of omics data. With growing interest in the use of such methods in computational metabolomics, we see an increased interest in and more publications on this issue. Moreover, machine learning, deep learning, and other AI-based methods are being tested, used, developed, and applied in metabolomics from data analysis to results interpretation. Several examples of cloud computing tools have been developed by and are available for the community. Finally we see more usage of workflows for data analysis enhancing results reproducibly both in open source and within commercial software tools. Within this Metabolites Special issue topic, we aim to capture such advancements through original publications, opinions or reviews on the topic. A wide range of topics from cloud based data analysis solutions and workflows to computational and statistical machine learning approaches applied on metabolomics data or as an integration means across omics datasets are highly welcomed.

This Special Issue will also explore how ML is transforming metabolomics as a field. The issue will aim to highlight (i) how to effectively use ML to develop new tools and analysis capabilities, (ii) how to create new ML approaches that support the unique aspects of metabolomics data and workflows, and (iii) how ML use is advancing studies that utilize metabolomics datasets. Reviews and forward-looking contributions that highlight ML’s transformative potential are also invited. Contributions covering comparative studies of ML and non-ML approaches and how we, as a community, share benchmark problems and datasets that measure ML progress in metabolomics are welcome.

Dr. Reza Salek
Dr. Justin Van der Hooft
Prof. Dr. Soha Hassoun
Dr. Simon Rogers
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Metabolites is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (2 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

13 pages, 2991 KiB  
Article
An Open-Source Pipeline for Processing Direct Infusion Mass Spectrometry Data of the Human Plasma Metabolome
Metabolites 2022, 12(8), 768; https://doi.org/10.3390/metabo12080768 - 21 Aug 2022
Cited by 1 | Viewed by 1707
Abstract
Direct infusion mass spectrometry (DIMS) is growing in popularity as an effective method for the screening of biological samples in clinical metabolomics. Being quick to execute, DIMS generally requires special skills when interpreting the results of measurements. By inspecting the similarities between two-dimensional [...] Read more.
Direct infusion mass spectrometry (DIMS) is growing in popularity as an effective method for the screening of biological samples in clinical metabolomics. Being quick to execute, DIMS generally requires special skills when interpreting the results of measurements. By inspecting the similarities between two-dimensional electrospray ionization with quadrupole time-of-flight (ESI-QTOF) and matrix-assisted laser desorption/ionization (MALDI) mass spectra, the pipeline for processing QTOF mass spectra using open-source packages (MALDIquant, MSnbase and MetaboAnalystR) was tested. Previously, all algorithmic workflows have relied on the application of software either provided by a vendor or privately developed by enthusiasts. Here, we computationally examined two ways of interpreting the DIMS results of human blood metabolomic profiling. The studied spectra were acquired using ESI-QTOF maXis Impact II (Bruker Daltonics, Billerica, MA, USA), then pre-processed using COMPASS/DataAnalysis commercial software and mapped onto the metabolites using in-lab-developed MatLab scripts. Alternatively, in this work we used the open-source packages MALDIquant, for spectrum pre-processing, and MetaboAnalystR, for data interpretation, instead of the low-availability commercial and home-made tools. Using a set of 100 plasma samples (20 from volunteers with normal body mass index and 80 from patients at different stages of obesity), we observed a high degree of concordance in annotated metabolic pathways between the proprietary DataAnalysis/MatLab pipeline and our freely available solution. Full article
(This article belongs to the Special Issue Metabolomics in the Age of Cloud Computing, AI and Machine Learning)
Show Figures

Figure 1

23 pages, 3783 KiB  
Article
MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition
Metabolites 2022, 12(4), 276; https://doi.org/10.3390/metabo12040276 - 22 Mar 2022
Cited by 3 | Viewed by 2344
Abstract
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 [...] Read more.
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. Full article
(This article belongs to the Special Issue Metabolomics in the Age of Cloud Computing, AI and Machine Learning)
Show Figures

Graphical abstract

Back to TopTop