Abstract
Cancer, an intricate field rooted in the study of abnormal cell growth and proliferation, presents bothfascination and complexity within biology. Unravelling the mysteries of cancer holds the key to addressing
numerous pressing issues in our society, from finding cures for debilitating malignancies to devising
safe and effective therapies and establishing guidelines for preventive measures. However, the intricate
nature of cancer biology has posed significant challenges to medical progress. The inherent variability
in experimental replication has contributed to less robust findings and many non-reproducible claims.
Recent advancements in computational technologies have paved the way for innovative mathematical
and computational techniques that promise to shed light on the intricate mechanisms underlying
cancer. Notably, the field of Systems Biology has emerged as a beacon of hope, offering novel methodologies
to tackle long-standing challenges in more efficient and resilient ways. Among these approaches,
genome-scale metabolic models (GSMM) are tools capable of simulating complex metabolic conditions
and bridging the gap between molecular alterations at the cellular level and the broader systemic
impacts on organisms. This thesis dives into the potential of GSMMs in the context of cancer research
and explores their applications through focused case studies while also integrating cutting-edge
machine-learning techniques. These machine-learning methodologies are engineered to uncover hidden
patterns within cancer-related data, offering insights that can pave the way for groundbreaking discoveries
or illuminate directions for future medical investigations. Modern computational techniques
are fundamental in advancing cancer biology research. Our study showcases the potential of GSMMs
within cancer research (e.g. Radiogenomics and Multi-omics), particularly in precision oncology. By
synergising GSMMs with advanced multi-modal machine learning, I illuminate the potential of this fusion
to reshape the landscape of cancer research, ultimately leading to more personalised and effective
strategies for understanding and combating this complex disease.
| Date of Award | 30 Sept 2024 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Yifeng Zeng (Supervisor), Annalisa Occhipinti (Supervisor) & Claudio Angione (Supervisor) |