Advancing systems biology of yeast through machine learning and comparative genomics

Aug 5, 2026

Upload by Anda

Synthetic biology enables production of valuable commodities, pharmaceuticals, and chemicals, with yeasts such as Saccharomyces cerevisiae serving as widely used microbial cell factories. Yet, detailed cellular metabolism and physiological properties remain insufficiently characterized for many yeast species. This thesis leverages large-scale yeast data using state-of-the-art machine learning and comparative genomics to elucidate yeast traits and metabolism. Machine learning predicts gene essentiality and kcat, supports enzyme-constrained model reconstruction, and identifies drivers of protein production. A comparative genomics toolbox characterizes horizontal gene transfer and substrate-use mechanisms across yeast genomes.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages68

Reading Time2 h 51 min

AccessFree to download

Abstract

Machine learning for yeast traits

Enzyme turnover prediction and enzyme-constrained models

Protein production feature analysis

Comparative genomics and HGT toolbox

THESIS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Advancing systems biology of yeast through machine learning and comparative genomics LE YUAN Division of Systems and Synthetic Biology Department of Life Sciences CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2023 Advancing systems biology of yeast through machine learning and comparative genomics LE YUAN ISBN 978-91-7905-818-0 © Le Yuan, 2023. Doktorsavhandlingar vid Chalmers tekniska högskola Ny serie nr 5284 ISSN 0346-718X Division of Systems and Synthetic Biology Department of Life Sciences Chalmers University of Technology SE-412 96 Gothenburg Sweden Telephone + 46 (0)31-772 1000 Cover illustration: Using machine learning techniques and comparative genomics to gain a deeper insight into gene essentiality, enzyme design and genome evolution of yeasts. Printed by Chalmers digitaltryck Gothenburg, Sweden 2023 Advancing systems biology of yeast through machine learning and comparative genomics Le Yuan Department of Life Sciences Chalmers University of Technology Abstract Synthetic biology has played a pivotal role in accomplishing the production of high value commodities, pharmaceuticals, and bulk chemicals. Fueled by the breakthrough of synthetic biology and metabolic engineering, Saccharomyces cerevisiae and various other yeasts (such as Yarrowia lipolytica, Pichia pastoris) have been proven to be promising microbial cell factories and are frequently used in scientific studies. However, the cellular metabolism and physiological properties for most of the yeast species have not been characterized in detail. To address these knowledge gaps, this thesis aims to leverage the large amounts of data available for yeast species and use state-of-the-art machine learning techniques and comparative genomic analysis to gain a deeper insight into yeast traits and metabolism. In this thesis, machine learning was applied to various unresolved biological problems on yeasts, i.e., gene essentiality, enzyme turnover number (kcat), and protein production. In the first part of the work, machine learning approaches were employed to predict gene essentiality based on sequence features and evolutionary features. It was demonstrated that the essential gene prediction could be substantially improved by integrating evolutionbased features. Secondly, a high-quality deep learning model DLKcat was developed to predict kcat values by combining a graph neural network for substrates and a convolutional neural network for proteins. By predicting kcat profiles for 343 yeast/fungi species, enzymeconstrained models were reconstructed and used to further elucidate the cellular metabolism on a large scale. Lastly, a random forest algorithm was adopted to investigate feature importance analysis on protein production, it was found that post-translational modifications (PTMs) have a relatively higher impact on protein production compared with amino acid composition. In comparative genomics, a comprehensive toolbox HGTphyloDetect was developed to facilitate the identification of horizontal gene transfer (HGT) events. Case studies on some yeast species demonstrated the ability of HGTphyloDetect to identify horizontally acquired genes with high accuracy. In addition, through systematic evolution analysis (e.g., HGT, gene family expansion) and genome-scale metabolic model simulation, the underlying mechanisms for substrate utilization were further probed across large-scale yeast species. Keywords: machine learning, deep learning, gene essentiality, enzyme turnover number, horizontal gene transfer, yeast species List of Publications This thesis is based on the following publications: Paper I: Hongzhong Lu†, Feiran Li†, Le Yuan†, Iván Domenzain, Rosemary Yu, Hao Wang, Gang Li, Yu Chen, Boyang Ji, Eduard J Kerkhoven, Jens Nielsen. Yeast metabolic innovations emerged via expanded metabolic network and gene positive selection. Molecular Systems Biology 17.10 (2021): e

FAQs

What problems in yeast biology does the thesis address using machine learning?

It targets unresolved issues including gene essentiality, enzyme turnover number (kcat), and protein production in yeasts.

How is kcat predicted in the thesis?

A deep learning model (DLKcat) combines a graph neural network for substrates with a convolutional neural network for proteins to predict kcat values and kcat profiles across many yeast/fungi species.

What role does comparative genomics play in the thesis?

It introduces a toolbox, HGTphyloDetect, to identify horizontal gene transfer events and uses additional evolutionary analyses and genome-scale metabolic model simulations to investigate substrate utilization mechanisms.

Aug 5, 2026

Upload by Anda