Investigating Ageing-Associated DNA Methylation Changes in Allosomes
Gerber, Lauren Jaye (2023)
2023
Master's Programme in Biomedical Technology
Lääketieteen ja terveysteknologian tiedekunta - Faculty of Medicine and Health Technology
This publication is copyrighted. Only for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-05-09
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202304264707
https://urn.fi/URN:NBN:fi:tuni-202304264707
Tiivistelmä
Background: While several studies have examined ageing and deoxyribonucleic acid (DNA) methylation at the autosomal level, it has been common practice to omit allosomes, or sex chromosomes, from such analyses. The drawback of using this approach is that interesting allosomal findings could be missed. As a result, allosomes were included in this thesis to assess if they were integral to ageing in both women and men. Specifically, DNA methylation sites, or cytosine-phosphate-guanine (CpG) sites, were interrogated, as DNA methylation changes have been known to be implicated in ageing.
Materials and Methods: This thesis analysed DNA methylation patterns in the female X chromosome (FX), male X chromosome (MX), and male Y chromosome (MY) using 2314 participants, ranging in age from 18 to 76 years old. This was done using the Young Finn Study (YFS) and Grady Trauma Project (GTP) datasets. The aim was to identify statistically significant DNA methylation changes associated with ageing, or differentially methylated positions (DMPs). To do so, a linear regression model was used, which generated several DMPs, where their adjusted p-values were less than the 0.05 α-level. DMPs found across both datasets were considered most important given their potential biological significance. If DMPs were in coding regions of the genome, they were then mapped to their corresponding genes, or differentially methylated genes (DMGs). DMGs containing the highest number of DMPs were further investigated, both for their baseline biological mechanisms and relevance to ageing. As the linear regression analysis was only performed on allosomes, a co-expression tool was run to identify co-expressed genes across all chromosomes. GO term analysis was then performed on the list of top co-expressed genes to find associated biological pathways (BPs).
Results: Intersecting DMPs amounted to 486 for the FX, 805 for the MX, 13 for the MY, and 139 for the FX and MX intersection (FX ∩ MX). Intersecting DMGs totalled 330 for the FX, 421 for the MX, 7 for the MY, and 222 for the FX ∩ MX. While there was considerable overlap between the YFS and GTP datasets, the GTP dataset, which was comprised of fewer subjects but had a wider age range (AR), consistently yielded more detected DMPs for all chromosomal comparisons than the YFS dataset. In both datasets, a trend toward hypomethylation was observed in the FX and MX, but no conclusive evidence could be drawn regarding a trend for the MY’s direction of methylation. The DMGs containing the highest numbers of DMPs were BCOR for the FX and MX and NLGN4Y for the MY, and these results were comparable to previously published results. Interestingly, GO term BPs yielded a high concentration of pathways involved in ribonucleic acid (RNA) splicing and processing.
Conclusions: The 3 main objectives of this thesis were met: 1) Finding a list of DMPs and DMGs shared across both datasets 2) Evaluating how these findings compared to previously published results 3) Performing co-expression and GO term BP analysis on the thesis findings – a step not taken in previous studies. Despite differences in the pipeline, datasets, and BeadChips, there was considerable overlap between this thesis and the previously published results. Further research needs to be performed to determine if allosomes are enriched for RNA splicing and processing, and what that may mean about the interplay of DNA methylation at both transcriptional and epigenetic levels.
Materials and Methods: This thesis analysed DNA methylation patterns in the female X chromosome (FX), male X chromosome (MX), and male Y chromosome (MY) using 2314 participants, ranging in age from 18 to 76 years old. This was done using the Young Finn Study (YFS) and Grady Trauma Project (GTP) datasets. The aim was to identify statistically significant DNA methylation changes associated with ageing, or differentially methylated positions (DMPs). To do so, a linear regression model was used, which generated several DMPs, where their adjusted p-values were less than the 0.05 α-level. DMPs found across both datasets were considered most important given their potential biological significance. If DMPs were in coding regions of the genome, they were then mapped to their corresponding genes, or differentially methylated genes (DMGs). DMGs containing the highest number of DMPs were further investigated, both for their baseline biological mechanisms and relevance to ageing. As the linear regression analysis was only performed on allosomes, a co-expression tool was run to identify co-expressed genes across all chromosomes. GO term analysis was then performed on the list of top co-expressed genes to find associated biological pathways (BPs).
Results: Intersecting DMPs amounted to 486 for the FX, 805 for the MX, 13 for the MY, and 139 for the FX and MX intersection (FX ∩ MX). Intersecting DMGs totalled 330 for the FX, 421 for the MX, 7 for the MY, and 222 for the FX ∩ MX. While there was considerable overlap between the YFS and GTP datasets, the GTP dataset, which was comprised of fewer subjects but had a wider age range (AR), consistently yielded more detected DMPs for all chromosomal comparisons than the YFS dataset. In both datasets, a trend toward hypomethylation was observed in the FX and MX, but no conclusive evidence could be drawn regarding a trend for the MY’s direction of methylation. The DMGs containing the highest numbers of DMPs were BCOR for the FX and MX and NLGN4Y for the MY, and these results were comparable to previously published results. Interestingly, GO term BPs yielded a high concentration of pathways involved in ribonucleic acid (RNA) splicing and processing.
Conclusions: The 3 main objectives of this thesis were met: 1) Finding a list of DMPs and DMGs shared across both datasets 2) Evaluating how these findings compared to previously published results 3) Performing co-expression and GO term BP analysis on the thesis findings – a step not taken in previous studies. Despite differences in the pipeline, datasets, and BeadChips, there was considerable overlap between this thesis and the previously published results. Further research needs to be performed to determine if allosomes are enriched for RNA splicing and processing, and what that may mean about the interplay of DNA methylation at both transcriptional and epigenetic levels.