Cover Image

Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin

Approximately 2.3 million women were diagnosed with breast cancer (BC) in 2020 and nearly 30% of women diagnosed with early-stage breast cancer will later develop metastatic disease. Despite the development and discovery of drugs and pharmacotherapy for breast cancer, the 5-year survival rate for pe...

Full description

Saved in:
Bibliographic Details
Main Author: Jalaludin, Nadia
Format: Thesis
Language:English
Published: 2023
Subjects:
Cancer
Online Access:https://ir.uitm.edu.my/id/eprint/84271/1/84271.pdf
https://ir.uitm.edu.my/id/eprint/84271/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uitm.ir.84271
record_format eprints
spelling my.uitm.ir.842712024-07-25T11:18:49Z https://ir.uitm.edu.my/id/eprint/84271/ Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin Jalaludin, Nadia Cancer Approximately 2.3 million women were diagnosed with breast cancer (BC) in 2020 and nearly 30% of women diagnosed with early-stage breast cancer will later develop metastatic disease. Despite the development and discovery of drugs and pharmacotherapy for breast cancer, the 5-year survival rate for people with metastatic breast cancer (MBC) remains low. Therefore, the objective of this study is to: a) mine and integrate clinical, phenotype and genotype data that contributes to the occurrence of MBC, b) build a prediction model that can predict possibility of occurrence to metastatic state of breast cancer based on factors previously determined in (a), and c) to validate findings from (a) and (b) through systematic review of randomised controlled trials of MBC. For objective (a), genotype and clinical data was mined from databases such as cBioportal and Genomic Data Common (GDC) portal, and was analysed using principal component analysis (PCA; after feature selection) and multiple correspondence analysis (MCA) in R. The data was then subjected to subsequent pathway mapping, Gene Ontology (GO) mapping and protein-protein interaction (PPI) to investigate its connection to the metastatic phenotype. The odds ratio of mutated gene, disease similarities and hierarchical clustering were also done before all the result was consolidated. For objective (b), prediction model was generated based on the outcome of (a) by using the Random Forest (RF) algorithm and validated by 5-fold cross validation. Additionally, the sensitivity and specificity of each model was also calculated. Meanwhile, for objective (c), six keywords: “metastatic breast cancer, chemotherapy, hormonal therapy, targeted therapy, gene and progression free survival” were used in PubMed, Scopus, Web of Science, Cochrane and Science Direct to further validate the previous findings. Based on all of these evaluations, the findings suggest that mRNA and genetic profiling can differentiate between breast cancer and metastatic breast cancer patients and more attention should be paid to YAP1 and SP7 genes. 2023 Thesis NonPeerReviewed text en https://ir.uitm.edu.my/id/eprint/84271/1/84271.pdf Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin. (2023) PhD thesis, thesis, Universiti Teknologi MARA (UiTM). <http://terminalib.uitm.edu.my/84271.pdf>
institution Universiti Teknologi Mara
building Tun Abdul Razak Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Mara
content_source UiTM Institutional Repository
url_provider http://ir.uitm.edu.my/
language English
topic Cancer
spellingShingle Cancer
Jalaludin, Nadia
Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
description Approximately 2.3 million women were diagnosed with breast cancer (BC) in 2020 and nearly 30% of women diagnosed with early-stage breast cancer will later develop metastatic disease. Despite the development and discovery of drugs and pharmacotherapy for breast cancer, the 5-year survival rate for people with metastatic breast cancer (MBC) remains low. Therefore, the objective of this study is to: a) mine and integrate clinical, phenotype and genotype data that contributes to the occurrence of MBC, b) build a prediction model that can predict possibility of occurrence to metastatic state of breast cancer based on factors previously determined in (a), and c) to validate findings from (a) and (b) through systematic review of randomised controlled trials of MBC. For objective (a), genotype and clinical data was mined from databases such as cBioportal and Genomic Data Common (GDC) portal, and was analysed using principal component analysis (PCA; after feature selection) and multiple correspondence analysis (MCA) in R. The data was then subjected to subsequent pathway mapping, Gene Ontology (GO) mapping and protein-protein interaction (PPI) to investigate its connection to the metastatic phenotype. The odds ratio of mutated gene, disease similarities and hierarchical clustering were also done before all the result was consolidated. For objective (b), prediction model was generated based on the outcome of (a) by using the Random Forest (RF) algorithm and validated by 5-fold cross validation. Additionally, the sensitivity and specificity of each model was also calculated. Meanwhile, for objective (c), six keywords: “metastatic breast cancer, chemotherapy, hormonal therapy, targeted therapy, gene and progression free survival” were used in PubMed, Scopus, Web of Science, Cochrane and Science Direct to further validate the previous findings. Based on all of these evaluations, the findings suggest that mRNA and genetic profiling can differentiate between breast cancer and metastatic breast cancer patients and more attention should be paid to YAP1 and SP7 genes.
format Thesis
author Jalaludin, Nadia
author_facet Jalaludin, Nadia
author_sort Jalaludin, Nadia
title Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
title_short Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
title_full Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
title_fullStr Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
title_full_unstemmed Understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / Nadia Jalaludin
title_sort understanding the occurrence of metastatic breast cancer through clinical, phenotype and genotype data, and the employment of machine learning / nadia jalaludin
publishDate 2023
url https://ir.uitm.edu.my/id/eprint/84271/1/84271.pdf
https://ir.uitm.edu.my/id/eprint/84271/
_version_ 1806422062905950208
score 13.211869

Similar Items