Back to search

The University of Edinburgh

Deciphering the role of DNA sequence features in human gene expression using factorial design of experiments

Abstract

dc:description.abstract

Gene expression is regulated at multiple levels, including the coding sequence itself. One key factor is guanine and cytosine (GC) content, which has been shown to enhance gene expression through various mechanisms. In the human genome, regions with distinct GC content, known as isochores, define functional domains with unique properties. However, the strong correlation between GC content (nucleotide level), CpG dinucleotide frequency, and codon optimality (trinucleotide level) makes it challenging to disentangle their individual contributions to gene expression. To address this, I used a synthetic library of synonymous Green Fluorescent Protein (GFP) variants generated via a full factorial design of experiments. Over 150 GFP variants were systematically generated by varying GC content, CpG dinucleotide frequency, and codon optimality at three levels (low, medium, and high). Transient overexpression experiments in two human cell lines revealed that GC content primarily determined steady-state mRNA levels, whereas codon optimality had a stronger influence on protein levels. Interestingly, CpG frequency had the least impact, largely due to its intrinsic negative correlation with codon optimality. Taken together, these sequence features accounted for the majority of the differences in gene expression observed. I also investigated the role of the Zinc-Finger Antiviral Protein (ZAP) and its cofactor TRIM25 in regulating GFP expression. While CpG-rich viral RNA is known to be suppressed by ZAP during the interferon response, depletion of endogenous ZAP and TRIM25 in human cells did not significantly affect CpG-rich GFP variants. Additionally, increased GFP expression imposed a cellular burden on cells, suggesting an interplay between sequence composition and cellular stress. To extend these findings, I constructed and validated a stable HeLa cell pool expressing GFP variants tagged with unique 20-nucleotide barcodes in the 3′ UTR. Using flow-seq, I analysed protein expression and the effects of gene silencing over prolonged passaging, confirming codon optimality as the most critical factor. Moreover, differences in gene silencing were dependent on the combination of coding sequence features. Overall, these findings highlight how different sequence features influence gene expression in human cells, with important implications for biotechnology, synthetic biology, and gene regulation.

Degree

thesis:*
Grantor dc:publisher
The University of Edinburgh
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Cano Aroca, Laura
Advisors dc:contributor.advisor
  • Kudla, Grzegorz
  • Sproul, Duncan

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
University of Edinburgh
Base URL
era.ed.ac.uk/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Cano Aroca, Laura. Deciphering the role of DNA sequence features in human gene expression using factorial design of experiments. The University of Edinburgh, 2025. https://hdl.handle.net/1842/44176