Mikhail G. Dozmorov, PhD

The human genome is non-random—DNA from each chromosome is folded into the highly organized three-dimensional (3D) structure. The three-dimensional (3D) structure has emerged as a higher-order regulatory layer orchestrating cell type-specific gene expression and other cellular processes. Abnormal changes in the 3D structure of the genome are now a well-established hallmark of cancer. However, their role in cancer metastasis, the primary cause of death, is unknown. This lack of understanding is particularly exacerbated for triple-negative breast cancer (TNBC) patients that currently lack treatment options at the metastatic stage of their disease. Comparing the 3D structure between primary and metastatic cancers will enable the detection of changes in the 3D structure that are associated with metastasis. Linking those changes with genes and cellular pathways will help to define druggable biomarkers that can be targeted to prevent metastasis. However, recently developed sequencing technologies for capturing the 3D structure of the genome remain imperfect, requiring proper normalization and statistical analysis of the data. This project will create biostatistical methods and bioinformatics software to maximize the detection of biologically relevant changes in the 3D structure of the genome. The methods will have broad application for defining statistically significant 3D changes in any experimental conditions. This project will compare primary and metastatic 3D structures obtained from patient-derived xenograft (PDX) models and identify metastasis-associated changes in the 3D genome. Using pharmacogenomics database and literature mining, genes associated with the 3D changes will be linked with the Food and Drug Administration (FDA)- approved drugs and prioritized for further drug screening in efforts to prevent metastasis.