Quantifying the driverness of copy number aberrations in prostate cancer
<p>Cancer evolves by a continuous process of selection, clonal expansion, and genetic diversification. The driver/passenger model distinguishes between driver mutations and passengers, and many driver identification methods have been developed. However, there are flaws in this model. The drive...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2024
|
| Subjects: |
| _version_ | 1826313345436745728 |
|---|---|
| author | Al-Muftah, NJ |
| author2 | Wedge, D |
| author_facet | Wedge, D Al-Muftah, NJ |
| author_sort | Al-Muftah, NJ |
| collection | OXFORD |
| description | <p>Cancer evolves by a continuous process of selection, clonal expansion, and genetic diversification. The driver/passenger model distinguishes between driver mutations and passengers, and many driver identification methods have been developed. However, there are flaws in this model. The driver/passenger model is dichotomous and does not reflect the gradient of driverness of somatic mutations. Moreover, p-values and recurrence may fail to detect lowly recurrent but important events in cancer. Lastly, the driverness of copy number aberrations (CNAs) in cancer have not been extensively studied. To overcome these limitations, I introduced novel methods to evaluate the driverness of CNAs in cancer. I applied them to CNA datasets of prostate cancer (PrCa) cohorts.</p>
<p>I developed a continuous measure of driverness based on effect sizes of somatic CNAs in a cohort of 159 primary PrCa samples, leading to the identification of novel positively and negatively selected CNA regions. I also revealed the relationship between the selection patterns observed in the CNAs to their biological role in PrCa. I characterised CNA drivers using functional impact scores of a range of genomic and regulatory features. In addition to confirming expected relationships such as enrichment for tumour suppressor genes in regions of copy number loss and oncogenes in gained regions, I made novel discoveries, including a cluster of regions subject to recurrent loss of heterozygosity enriched in AR binding sites and super enhancers, indicating the disruption of their corresponding signalling pathways during PrCa progression. I also developed a classifier of low grade vs. high grade PrCa using copy number profiles of patients. The model correctly classified 72.9% and 80.2% of high grade and low grade primary PrCa samples, respectively. In summary, I established a quantitative model of CNA driverness, changing the perception of driverness from binary to continuous. This work provides new insights on CNA driverness and their role in PrCa.</p>
|
| first_indexed | 2024-09-25T04:13:21Z |
| format | Thesis |
| id | oxford-uuid:155fa460-65ab-4f74-ba2b-74d323667409 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:13:21Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | oxford-uuid:155fa460-65ab-4f74-ba2b-74d3236674092024-07-10T10:08:15ZQuantifying the driverness of copy number aberrations in prostate cancerThesishttp://purl.org/coar/resource_type/c_db06uuid:155fa460-65ab-4f74-ba2b-74d323667409Cancer GenomicsBioinformaticsComputational biologyEnglishHyrax Deposit2024Al-Muftah, NJWedge, DWedge, DAnsari-Pour, NAnsari-Pour, NSchuster-Boeckler, P<p>Cancer evolves by a continuous process of selection, clonal expansion, and genetic diversification. The driver/passenger model distinguishes between driver mutations and passengers, and many driver identification methods have been developed. However, there are flaws in this model. The driver/passenger model is dichotomous and does not reflect the gradient of driverness of somatic mutations. Moreover, p-values and recurrence may fail to detect lowly recurrent but important events in cancer. Lastly, the driverness of copy number aberrations (CNAs) in cancer have not been extensively studied. To overcome these limitations, I introduced novel methods to evaluate the driverness of CNAs in cancer. I applied them to CNA datasets of prostate cancer (PrCa) cohorts.</p> <p>I developed a continuous measure of driverness based on effect sizes of somatic CNAs in a cohort of 159 primary PrCa samples, leading to the identification of novel positively and negatively selected CNA regions. I also revealed the relationship between the selection patterns observed in the CNAs to their biological role in PrCa. I characterised CNA drivers using functional impact scores of a range of genomic and regulatory features. In addition to confirming expected relationships such as enrichment for tumour suppressor genes in regions of copy number loss and oncogenes in gained regions, I made novel discoveries, including a cluster of regions subject to recurrent loss of heterozygosity enriched in AR binding sites and super enhancers, indicating the disruption of their corresponding signalling pathways during PrCa progression. I also developed a classifier of low grade vs. high grade PrCa using copy number profiles of patients. The model correctly classified 72.9% and 80.2% of high grade and low grade primary PrCa samples, respectively. In summary, I established a quantitative model of CNA driverness, changing the perception of driverness from binary to continuous. This work provides new insights on CNA driverness and their role in PrCa.</p> |
| spellingShingle | Cancer Genomics Bioinformatics Computational biology Al-Muftah, NJ Quantifying the driverness of copy number aberrations in prostate cancer |
| title | Quantifying the driverness of copy number aberrations in prostate cancer |
| title_full | Quantifying the driverness of copy number aberrations in prostate cancer |
| title_fullStr | Quantifying the driverness of copy number aberrations in prostate cancer |
| title_full_unstemmed | Quantifying the driverness of copy number aberrations in prostate cancer |
| title_short | Quantifying the driverness of copy number aberrations in prostate cancer |
| title_sort | quantifying the driverness of copy number aberrations in prostate cancer |
| topic | Cancer Genomics Bioinformatics Computational biology |
| work_keys_str_mv | AT almuftahnj quantifyingthedrivernessofcopynumberaberrationsinprostatecancer |