Large-scale transcriptomic studies are among of the most comprehensive accounts we have of the biological processes underlying human brain development and ageing. However, many analyses and descriptive models applied to gene expression data implicitly assume that developmental change is continuous and uninterrupted. Perhaps this bias is often overlooked because the emphasis is on what is changing during development rather than how development itself is changing. Indeed, despite the richness of transcriptomic data and its capacity to recapitulate higher-order functions, few have used it to understand the dynamics of brain development. Gene expression is determined by complex, high-dimensional interactions of the gene regulatory network. Dynamic systems theory states that the interactions of components in any complex systems will converge on certain stable patterns, also known as attractor states. To approximate these stable states, the current study leveraged robust and sparse k-means clustering to identify tissue samples with similar patterns of gene expression across the transcriptome. Sample ages were then used to visualize when in developmental time these stable patterns are present. The resulting model describes the developmental dynamics of the brain transcriptome as a series of non-linear, overlapping states that progress across the lifespan. / Thesis / Master of Science (MSc)
| Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/26142 |
| Date | January 2021 |
| Creators | Arbabi, Keon |
| Contributors | Kathryn, Murphy, Neuroscience |
| Source Sets | McMaster University |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0019 seconds