The ultimate goal of cancer biology is to enable the use of molecular profiles, such as DNA, RNA, protein and epigenetic data, of individual cancer patients for diagnosis and treatment. There is substantial need for better ways to choose chemotherapy drugs in individual cancer patients. For example, patients over 65 with acute myeloid leukemia (AML), an aggressive blood cancer, have no better prognosis today than they did in 1980. For a growing number of diseases, there is a fair amount of data on molecular profiles from patients. The most important step necessary to realize the ultimate goal is to identify molecular markers (e.g., certain genes or mutations) in these molecular data that predict treatment outcomes, such as response to each chemotherapy drug. However, due to the high-dimensionality (i.e., the number of variables is much greater than the number of samples) along with potential biological or experimental confounders, it is an open challenge to identify robust biomarkers that are replicated across different studies.

In this talk, I will present two distinct machine learning approaches to resolve this challenge. These methods learn the low-dimensional features that are likely to represent important molecular events in the disease process in an unsupervised fashion, based on molecular profiles from multiple populations of cancer patients. I will present two applications of these two methods – AML, and ovarian cancer. When the first method was applied to AML data in collaboration with UW Center for Cancer Innovation, a novel molecular marker for topoisomerase inhibitors, widely used chemotherapy drugs in AML treatment, was revealed.  The other method applied to ovarian cancer data led to a potential molecular driver for an aggressive ‘mesenchymal’ subtype, in collaboration with UW Pathology and UW Genome Sciences. These findings will lead to better ways to treat cancer.

This research is funded by the American Cancer Society, the National Science Foundation, and the National Institutes of Health.

Current diagnostic tests for diffuse large B-cell lymphoma use the updated WHO criteria based on biologic, morphologic, and clinical heterogeneity. We have recently proposed a refined classification system based on subset-specific B-cell–associated gene signatures (BAGS) in the normal B-cell hierarchy which refines the ABC/GCB classification in DLBCL. The expression based BAGS-classification system added independent, significant prognostic information to the molecular ABC/GCB subclasses and the clinical International Prognostic Index (IPI). Additionally, the BAGS subtypes had distinct genetic profiles, differentially activated signaling pathways and specific responses to chemotherapeutic drugs used in a routine clinical setting for treatment of DLBCL.  During the presentation we will review the bioinformatics work behind the BAGS-classifier and give current status on the recent developments on differential network expression analysis between BAGS subtypes, detection of normal contamination and identification of clones with varying cellular origin. Finally, we want to discuss the development of a diagnostic platform for BAGS-assigned subtypes in B-cell malignancies, which may allow pathogenetic studies to improve disease management.

Estrogen receptor-α (ERα) is the defining and driving transcription factor in the majority of breast cancers (ERα+). Researchers in Jason Carroll’s lab have been exploring the binding dynamics of ERα and related co-factors using a variety of high-throughput genomic assays. Identifying significantly differentially bound sites reveals transcription factor binding patterns that are associated with tamoxifen resistance and clinical outcome (Ross-Innes 2012), as well as identifying related pioneer factors FoxA1 and GATA3 (Theodorou 2013). Recent work (Mohammed 2015) examines the interaction of ERα and the progesterone receptor (PR), showing that PR is not merely an ERα-induced gene target, but a driver of distinct regulatory programs. In the presence of progesterone PR associates with ERα to direct chromatin-binding events, resulting in a gene expression profile that is anti-proliferative and associated with good clinical outcome. Progesterone inhibits estrogen-mediated growth of ERα+ cell line xenografts and primary ERα+ breast tumor explants, and has increased anti-proliferative effects when coupled with an ERα antagonist (tamoxifen). Copy number loss of the gene coding for PR is associated with poor outcome in a large patient cohort. Together, these findings have important implications for prognosis and therapeutic intervention based on PR activity.

As a computational biologist, I will highlight analytical challenges of this research, particularly in the integrative analysis of proteomic (RIME), ChIP-seq, RNA-seq, and CNV data from patients and cell lines, assayed directly and in xenografts. In particular, I will discuss the need for computational assessment of the uniqueness of prognostic gene signatures derived from high-throughput genomic assays.

Cancer development is a highly complex evolutionary process driven by the dynamic feedback between somatic mutations and the organ microenvironment in which the tumor initiates and progresses. Spatial and temporal gradients of microenvironmental factors (e.g. nutrients, oxygen, and drug) result in selective pressures that affect the fitness of cancer cells and influence the overall composition of the tumor. It is imperative to capture the heterogeneity of a tumor across scales (of size and time) to achieve a deeper understanding of the evolutionary dynamics of a tumor and the ability to successfully intervene therapeutically. Recently, we developed a high-content imaged based method that is capable of multiplex phenotypic profiling of cell populations in response to co-occurring environmental selective pressures. This platform has been applied to both 2D monolayers and 3D patient-derived organoids to track changes in cell/organoid growth and morphology features under various environmental perturbations. Here we demonstrate that the local tumor microenvironment can influence the magnitude of the selective effect, with larger and more numerous organoids forming under physiologically relevant oxygen culture conditions (5% and 1% O2) compared to standard laboratory conditions (21% O2). Our image-based approach provides a deeper insight into the evolving cellular dynamics of a tumor, and utilizes reduced reagents and time, offering advantages over traditional biological assays.  

A cell’s stimulus response depends not only on the internal concentrations of signaling molecules but also on their spatial dynamics. Thus we need techniques to measure the bulk kinetics of molecules in live cells to better identify mechanisms responsible for cell behavior. In this talk I will present a new method called multimodal Image Correlation Spectroscopy (mICS) that estimates ballistic and anisotropic diffusive molecular motion from live cell microscopy. I will demonstrate the utility of mICS as applied to a prostate cancer (PCa) cell line system. Androgen receptor (AR) is the key PCa signaling molecule that drives proliferation by translocating from cytoplasm to nucleus under ligand stimulation. This signal is the main therapeutic target in early stage PCa. mICS analysis shows that AR undergoes diffusion, confined movement, and bound motion following ligand stimulation. Together with and simulations from a simple computational model, these new findings imply that cytoplasmic AR translocation is dominated by diffusion rather than microtubule transport. This has implications for therapeutic strategy.

A central challenge in cancer therapeutics is resistance -- ‘primary / innate’ as well as ‘secondary / acquired’ -- manifested in any of a variety of forms to almost every mode of treatment.  Understanding underlying causes of resistance and developing capabilities for addressing them more effectively requires comprehension from multi-variate systems perspective.  In this seminar, I will present a set of examples of recent work from our laboratory (in close collaboration with a number of others), which offer new insights being gained from application of integrated experimental/computational approaches to clinically-relevant challenges including tumor adaptation, tumor heterogeneity, and tumor microenvironment influence.