For our next CDSL Webinar we will have a guest lecture by Dr. Russell Rockne from Beckman Research Institute, City of Hope National Medical Center. Abstract: Temporal dynamics of gene expression inform
For our next CDSL Webinar we will have a guest lecture by Dr. Russell Rockne from Beckman Research Institute, City of Hope National Medical Center.
Abstract: Temporal dynamics of gene expression inform cellular and molecular perturbations associated with disease development and evolution. Given the complexity of high-dimensional temporal genomic data, an analytic framework guided by a robust theory is needed to interpret time-sequential changes and to predict system dynamics. Using a murine model of acute myeloid leukemia (AML) we model temporal dynamics of the transcriptome of peripheral blood mononuclear cells derived from time-sequential bulk RNA-seq (mRNA) and micro-RNA-seq (miRNA) expression data in a CBFB-MYH11 (CM) knock-in mouse model (Cbfb+/56M/Mx1-Cre; C57BL/6) that mimics human inv(16) AML. Blood was collected from both CM (n=7) and control mice lacking the transgene (n=7) before CM induction and every month for 10 months post-induction.
From the time-series mRNA and miRNA expression data, we construct an AML state-space with the singular value decomposition. Using the samples’ location in the state-space, we applied the state-transition model which views the development of AML as a particle undergoing Brownian motion in a potential with three states corresponding to critical points: health (c1), unstable transition (c2), and overt AML (c3). The dynamics of mRNA and miRNA expression in the state-space relative to the critical points accurately predicted AML development in two validation studies (N=12, logrank p<0.01) and identified transcriptional perturbations associated with leukemia progression, including: cell signaling, inflammation, and metabolic pathways. Moreover, the geometry of the mRNA and miRNA state-spaces provided novel interpretations of gene dynamics, aligned gene signals that were not synchronized in time across mice, and provided quantifications of gene and pathway contributions to leukemia development. Interestingly, the acute angle between the mRNA and miRNA state-spaces revealed a mapping between related but distinctly different ‘epigenetic’ representations of AML. Our state-transition mathematical model and the geometry of the mRNA and miRNA state-spaces provides a theory-guided, insightful analysis of longitudinal multi-omic data which predicts leukemia progression and suggests novel targets for therapeutic interventions.
Dr. Rockne received his Ph.D. in Applied Mathematics at the University of Washington where he developed predictive mathematical models of brain cancer response to radiation therapy with PhD advisor Dr. Kristin Swanson. He then performed postdoctoral research at Northwestern University and was subsequently recruited to the Beckman Research Institute in California as an Assistant Professor, where he established the Division of Mathematical Oncology in the Department of Computational and Quantitative Medicine.
Dr. Rockne’s current research includes mathematical modeling as it relates to precision medicine, data science, computational systems biology, machine learning, and quantitative image analysis. Dr. Rockne is funded by the NCI, NINDS, the California Institute for Regenerative Medicine (CIRM), and is a PI in the Physical Sciences Oncology Network (PSON) and Cancer Systems Biology Consortium (CSBC).
Active areas of Dr. Rockne’s research include time-series genomic data analysis; modeling with single-cell sequencing data; and the integration of machine learning methods with mechanism-based mathematical models. Dr. Rockne is recognized as a leader in the field of Mathematical Oncology, with positions on several editorial boards, including JCO Clinical Cancer Informatics, and highly cited manuscripts and editorials, including a recent Roadmap article which outlines the next 5 years of research in the field of Mathematical Oncology.
Meeting LinkJoin ZoomGov Meeting
(Wednesday) 11:00 am - 12:00 pm
CDSLNCI CCR Cancer Data Science Lab