Translational bioinformatics in heritable cancer susceptibility diseases

Hereditary cancers, characterized by genetic anomalies like gene mutations that elevate cancer susceptibility, pose unique challenges. This thesis focuses on two such conditions: Fanconi anemia and retinoblastoma, each with distinct genetic causes and medical complexities. Fanconi anemia follows a Mendelian recessive pattern due to mutations in 23 identified genes in the Fanconi anemia DNA repair pathway. In contrast, retinoblastoma, the most common childhood eye cancer, primarily stems from RB1 tumor suppressor gene mutations. Fanconi anemia patients face an elevated risk of head and neck cancer due to DNA repair defects, necessitating alternative treatments. In retinoblastoma, germline RB1 mutations can lead to second primary malignancies. The unique nature of each disease presents specific medical and research challenges. Current Fanconi anemia research priorities include safer head and neck cancer therapies and understanding molecular mechanisms compensating for DNA repair deficiency. In retinoblastoma, research explores tumor subtype classification, biopsy-based assessments, and early second primary malignancy detection. Bioinformatics, bridging biology, computer science, and statistics, plays a crucial role in generating, processing, and analyzing OMICs data. This thesis employs various genomic bioinformatic approaches to study hereditary cancer susceptibility diseases, specifically Fanconi anemia and retinoblastoma. Chapter 2 identifies potential cancer vulnerability targets in FA-HNSCC cells with 11q22 amplifications through integrated genomics analysis. BIRC2-BIRC3 genes appear as promising therapeutic targets, as they regulate cell survival and death. Chapter 3 uses multi-OMICs analysis to identify gene signatures related to hypermethylation, differentiating MYCN-high RB1-proficient retinoblastomas. These signatures are functionally associated with various biological processes. Chapter 4 confirms radiogenomics findings by correlating MRI features with gene expression data in retinoblastoma cohorts. Significant associations between MRI features and photoreceptorness-expression signature are verified, indicating the applicability of radiogenomics in diagnostics. Chapter 5 conducts a whole-genome siRNA screen in FA-HNSCC cells, identifying RBBP9 as an essential gene with potential targetable vulnerability. Targeted therapies hold promise for the treatment of FA-HNSCCs. Chapter 6 characterizes structural variations and transcriptomic components related to ICL resistance in FA-deficient lymphocytic cell lines. The research highlights the complex dynamics of adaptation to DNA cross-linkers in FA-deficient cells. In summary, chapters 2-6 showcase translational bioinformatics approaches to address research challenges in hereditary cancer susceptibility diseases. Artificial intelligence methodologies and publicly available data play key roles in shaping the future of translational bioinformatics, offering hope for improved care for patients affected by rare cancers when combined with both wet-lab and dry-lab techniques and data sharing efforts.