Browsing by Subject "Tumor heterogeneity"

Now showing 1 - 3 of 3

Bayesian nonparametric models for biomedical data analysis
(2017-08) Zhou, Tianjian, Ph. D.; Müller, Peter, 1963 August 9-; Daniels, Michael; Ji, Yuan; Williamson, Sinead
In this dissertation, we develop nonparametric Bayesian models for biomedical data analysis. In particular, we focus on inference for tumor heterogeneity and inference for missing data. First, we present a Bayesian feature allocation model for tumor subclone reconstruction using mutation pairs. The key innovation lies in the use of short reads mapped to pairs of proximal single nucleotide variants (SNVs). In contrast, most existing methods use only marginal reads for unpaired SNVs. In the same context of using mutation pairs, in order to recover the phylogenetic relationship of subclones, we then develop a Bayesian treed feature allocation model. In contrast to commonly used feature allocation models, we allow the latent features to be dependent, using a tree structure to introduce dependence. Finally, we propose a nonparametric Bayesian approach to monotone missing data in longitudinal studies with non-ignorable missingness. In contrast to most existing methods, our method allow for incorporating information from auxiliary covariates and is able to capture complex structures among the response, missingness and auxiliary covariates. Our models are validated through simulation studies and are applied to real-world biomedical datasets.
Development of functionalized lineage tracing tools to characterize and manipulate complex biological systems
(2019-09-23) Al'Khafaji, Aziz Muhsin; Brock, Amy; Vishwanath, Iyer; Marcotte, Edward; Barrick, Jeffrey; Dalby, Kevin
Multicellular organisms are composed of heterogeneous groups of cells that, through specialized roles, work in a cooperative fashion to assume higher level functions. Flaws can arise in these multicellular systems, causing a breakdown of cooperativity. This is exemplified in cancer, where dysregulation of cellular proliferation gives rise to invasive tumors that may kill the host. This breakdown from regular order is self-perpetuating and culminates in variegating cell populations that accrue variation at the genotypic and phenotypic levels. Clonal diversity is a hallmark of cancer and a primary determinant in cancer’s ability to evade therapeutic treatment. Characterizing these heterogeneous evolving populations is a daunting challenge, as measurements need to resolve both the variation and clonal identity between constituent cells. Recent technological advancements have made these observations possible, with single-cell omics technologies driving granular resolution of heterogeneous populations and lineage tracing methods providing highly sensitive measurements of clonal composition. The work described in this text details the development and validation of a novel lineage tracing technology, Control Of Lineage by Barcode Enabled Recombinant Transcription (COLBERT), to tag, track, and manipulate clones within a complex biological system. We show that this technique is highly sensitive and broadly effective in multiple cell types. Further, we integrate COLBERT with other methodologies and single-cell workflows to characterize evolutionary dynamics of therapeutic resistance in a CLL model system. Implementing these combined approaches has uniquely enabled the identification of pre-existing drug tolerant subpopulations which are distinct in their clonal composition. With knowledge of these clonally linked phenotypes, we have been able to isolate high tolerance clones for further molecular characterization to uncover the factors stabilizing this survival phenotype. As our technical capabilities increase and our biological questions become more ambitious, it is clear that the next frontiers will be to simultaneously measure and integrate multiple facets of biology at single-cell resolution. The capacity to then manipulate these systems based on understandings of their phenotypic and clonal contexts will offer tremendous opportunities for biological engineering and discovery.
Quantitative characterization of tumor heterogeneity in breast cancer in vivo
(2020-09-23) Syed, Anum Kamal; Yankeelov, Thomas E.; Sorace, Anna G; Brock, Amy; Bankson, James; Virostko, John
Tumor heterogeneity provides a major challenge for the clinical treatment of breast cancer and is associated with poor patient prognosis and treatment failure. Heterogeneity of the tumor microenvironment and cancer cell phenotypes can result in nonuniform drug delivery and varying tumor response. The purpose of this dissertation is to develop novel methods for the quantitative characterization of tumor heterogeneity using noninvasive imaging, and then employ these methods to measure longitudinal changes in intratumoral heterogeneity in response to treatment. This objective is addressed in three parts. First, in a murine xenograft model of HER2+ breast cancer, we quantify temporal alterations of vascular, cellular, and hypoxic heterogeneity in response to the HER2 targeted therapy, trastuzumab. The results indicate that trastuzumab induces longitudinal increases in cellular and vascular heterogeneity, and longitudinal decreases hypoxic heterogeneity. Second, using multiparametric MRI data, we identify physiologically-distinct tumor habitats in two preclinical models of breast cancer, and measure the changes in tumor composition over time, in response to targeted or cytotoxic therapy. We identify tumor habitats associated with treatment response in both models of breast cancer and provide biological validation of the identified habitats using immunohistochemistry. Finally, in an effort to elucidate intertumoral heterogeneity, we utilize tumor habitats to classify HER2+ xenograft tumors into two tumor imaging phenotypes at baseline and measure the longitudinal response of each phenotype to targeted and cytotoxic therapies. The two tumor phenotypes progress and respond to therapeutic intervention differently, as one phenotype demonstrates more “therapy-sensitive” behavior, with significant decreases in tumor volume and increases in low-vascularity low-cellularity habitats in response to treatment. Overall, our results offer a novel methodology to quantify the spatiotemporal alterations in intratumoral heterogeneity in response to therapeutic intervention using clinically-translatable imaging technologies.