Machine Learning in Metabolic Diseases

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

Deadline for manuscript submissions: closed (15 November 2024) | Viewed by 6661

Special Issue Editor


E-Mail Website
Guest Editor
1. Research Unit of Molecular Epidemiology, Institute of Epidemiology, Helmholtz Zentrum München, Germany
2. Research Center for Environmental Health, 85764 Neuherberg, Germany
Interests: type 2 diabetes and complications; bioinformatics; machine learning; omics data integration; translational research

Special Issue Information

Dear Colleagues,

Machine learning (ML) concerns computer algorithms that improve their performance by learning from large sets of data. As a subdiscipline of artificial intelligence, ML has been developed and applied in analyzing complex data such as metabolomics to predict, identify and validate biomarkers / risk factors of metabolic diseases. The key steps of ML includes 1) data gathering and pre-processing; 2) model selection, training and testing; and 3) prediction, inference and applications. Large and high quality data enable good performance for predicting disease risk to develop efficient personalized diagnosis and therapy.

This Special Issue focuses on ML in metabolic diseases. Topics include studies aimed at developing and / or using ML in the following areas:

  • Collection of data (e.g., human cohort studies, clinical studies, biobanks), and data pre-processing (e.g., harmonization / normalization of individuals molecular profiles or clinical phenotypes);
  • Techniques for optimized ML model selection. ML methods may include supervised (e.g., regression and classification analysis, support vector machine and random forest) and unsupervised (e.g., clustering, principal component analysis, autoencoders and generative adversarial networks);
  • Application of ML for improved prediction, identification and validation of risk factors, modifiers and / or biomarkers of metabolic diseases.

Dr. Rui Wang-Sattler
Guest Editor

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.

Keywords

  • machine learning
  • supervised
  • unsupervised
  • model selection
  • training and testing
  • data pre-processing
  • prediction
  • identification
  • validation risk factors/biomarkers
  • metabolic disease

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (3 papers)

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

Research

17 pages, 2836 KiB  
Article
Intra-Individual Variations in How Insulin Sensitivity Responds to Long-Term Exercise: Predictions by Machine Learning Based on Large-Scale Serum Proteomics
by Jonas Krag Viken, Thomas Olsen, Christian André Drevon, Marit Hjorth, Kåre Inge Birkeland, Frode Norheim and Sindre Lee-Ødegård
Metabolites 2024, 14(6), 335; https://doi.org/10.3390/metabo14060335 - 15 Jun 2024
Viewed by 1647
Abstract
Physical activity is effective for preventing and treating type 2 diabetes, but some individuals do not achieve metabolic benefits from exercise (“non-responders”). We investigated non-responders in terms of insulin sensitivity changes following a 12-week supervised strength and endurance exercise program. We used a [...] Read more.
Physical activity is effective for preventing and treating type 2 diabetes, but some individuals do not achieve metabolic benefits from exercise (“non-responders”). We investigated non-responders in terms of insulin sensitivity changes following a 12-week supervised strength and endurance exercise program. We used a hyperinsulinaemic euglycaemic clamp to measure insulin sensitivity among 26 men aged 40–65, categorizing them into non-responders or responders based on their insulin sensitivity change scores. The exercise regimen included VO2max, muscle strength, whole-body MRI scans, muscle and fat biopsies, and serum samples. mRNA sequencing was performed on biopsies and Olink proteomics on serum samples. Non-responders showed more visceral and intramuscular fat and signs of dyslipidaemia and low-grade inflammation at baseline and did not improve in insulin sensitivity following exercise, although they showed gains in VO2max and muscle strength. Impaired IL6-JAK-STAT3 signalling in non-responders was suggested by serum proteomics analysis, and a baseline serum proteomic machine learning (ML) algorithm predicted insulin sensitivity responses with high accuracy, validated across two independent exercise cohorts. The ML model identified 30 serum proteins that could forecast exercise-induced insulin sensitivity changes. Full article
(This article belongs to the Special Issue Machine Learning in Metabolic Diseases)
Show Figures

Figure 1

18 pages, 15365 KiB  
Article
Prediction of Myocardial Infarction Using a Combined Generative Adversarial Network Model and Feature-Enhanced Loss Function
by Shixiang Yu, Siyu Han, Mengya Shi, Makoto Harada, Jianhong Ge, Xuening Li, Xiang Cai, Margit Heier, Gabi Karstenmüller, Karsten Suhre, Christian Gieger, Wolfgang Koenig, Wolfgang Rathmann, Annette Peters and Rui Wang-Sattler
Metabolites 2024, 14(5), 258; https://doi.org/10.3390/metabo14050258 - 30 Apr 2024
Viewed by 1252
Abstract
Accurate risk prediction for myocardial infarction (MI) is crucial for preventive strategies, given its significant impact on global mortality and morbidity. Here, we propose a novel deep-learning approach to enhance the prediction of incident MI cases by incorporating metabolomics alongside clinical risk factors. [...] Read more.
Accurate risk prediction for myocardial infarction (MI) is crucial for preventive strategies, given its significant impact on global mortality and morbidity. Here, we propose a novel deep-learning approach to enhance the prediction of incident MI cases by incorporating metabolomics alongside clinical risk factors. We utilized data from the KORA cohort, including the baseline S4 and follow-up F4 studies, consisting of 1454 participants without prior history of MI. The dataset comprised 19 clinical variables and 363 metabolites. Due to the imbalanced nature of the dataset (78 observed MI cases and 1376 non-MI individuals), we employed a generative adversarial network (GAN) model to generate new incident cases, augmenting the dataset and improving feature representation. To predict MI, we further utilized multi-layer perceptron (MLP) models in conjunction with the synthetic minority oversampling technique (SMOTE) and edited nearest neighbor (ENN) methods to address overfitting and underfitting issues, particularly when dealing with imbalanced datasets. To enhance prediction accuracy, we propose a novel GAN for feature-enhanced (GFE) loss function. The GFE loss function resulted in an approximate 2% improvement in prediction accuracy, yielding a final accuracy of 70%. Furthermore, we evaluated the contribution of each clinical variable and metabolite to the predictive model and identified the 10 most significant variables, including glucose tolerance, sex, and physical activity. This is the first study to construct a deep-learning approach for producing 7-year MI predictions using the newly proposed loss function. Our findings demonstrate the promising potential of our technique in identifying novel biomarkers for MI prediction. Full article
(This article belongs to the Special Issue Machine Learning in Metabolic Diseases)
Show Figures

Graphical abstract

23 pages, 4246 KiB  
Article
Identification of Cancer Driver Genes by Integrating Multiomics Data with Graph Neural Networks
by Hongzhi Song, Chaoyi Yin, Zhuopeng Li, Ke Feng, Yangkun Cao, Yujie Gu and Huiyan Sun
Metabolites 2023, 13(3), 339; https://doi.org/10.3390/metabo13030339 - 24 Feb 2023
Cited by 6 | Viewed by 2777
Abstract
Cancer is a heterogeneous disease that is driven by the accumulation of both genetic and nongenetic alterations, so integrating multiomics data and extracting effective information from them is expected to be an effective way to predict cancer driver genes. In this paper, we [...] Read more.
Cancer is a heterogeneous disease that is driven by the accumulation of both genetic and nongenetic alterations, so integrating multiomics data and extracting effective information from them is expected to be an effective way to predict cancer driver genes. In this paper, we first generate comprehensive instructive features for each gene from genomic, epigenomic, transcriptomic levels together with protein–protein interaction (PPI)-networks-derived attributes and then propose a novel semisupervised deep graph learning framework GGraphSAGE to predict cancer driver genes according to the impact of the alterations on a biological system. When applied to eight tumor types, experimental results suggest that GGraphSAGE outperforms several state-of-the-art computational methods for driver genes identification. Moreover, it broadens our current understanding of cancer driver genes from multiomics level and identifies driver genes specific to the tumor type rather than pan-cancer. We expect GGraphSAGE to open new avenues in precision medicine and even further predict drivers for other complex diseases. Full article
(This article belongs to the Special Issue Machine Learning in Metabolic Diseases)
Show Figures

Figure 1

Back to TopTop