Javascript must be enabled for the correct page display

The Potential for Machine Learning in Multi-Omics-based Microbiome Research

van Sloten, Maxime (2023) The Potential for Machine Learning in Multi-Omics-based Microbiome Research. Master's Thesis / Essay, Biomedical Sciences.


Download (1MB) | Preview
[img] Text
Restricted to Registered users only

Download (140kB)


In the last few decades extensive research has been done on the human microbiome, which is the sum of all microorganisms that live within and on our body. While the microbiome itself is influenced by many factors, it in turn also plays a vital role in influencing the health and disease of human beings. It is therefore no surprise that an increasing amount of effort has been put in gathering data on the human microbiome. The development of high-throughput technologies has allowed for the development of longitudinally personalized multi-omics profiling. Due to the complicated web of interactions between the human microbiome and the host, it is extremely complicated to determine significant health-related associations with certainty. A possible solution that has gotten considerable interest from scientists in recent years has been the rapidly developing field of machine learning (ML) due to its potential for integrating large datasets, creating models and predicting phenotypes. ML can generally be subdivided into two groups; unsupervised learning (like PCA and PCoA) and supervised learning (like SVM and RF) with deep learning as a technique that can be both unsupervised or supervised. ML applications have had positive results in recent years, as can be seen in microbiome studies on antimicrobial resistance and cancer where ML strategies rivaled and/or eclipsed traditional analyses. However, concepts such as the curse of dimensionality, high-quality data and interpretability are

Item Type: Thesis (Master's Thesis / Essay)
Supervisor name: Harmsen, H.J.M.
Degree programme: Biomedical Sciences
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 03 Apr 2023 11:55
Last Modified: 03 Apr 2023 11:55

Actions (login required)

View Item View Item