The UNC Computational Medicine Program and the SOM Office of Research announce the winners of the second annual Computational Medicine Pilot Grant Awards.

The winning teams include:

“Uncovering function in long noncoding RNAs through a new paradigm in evolutionary analysis”

  • Co-PI: Daniel Schrider, Ph.D., Assistant Professor, Department of Genetics
  • Co-PI: J. Mauro Calabrese, Ph.D., Assistant Professor, Department of Pharmacology

This team’s research objective: The human genome expresses thousands of long noncoding RNAs (lncRNAs), RNA transcripts >200 nucleotides in length that do not code for protein. Case studies show that lncRNAs can harbor essential functions that when abrogated give rise to disease. Accordingly, lncRNAs represent a new class of possible biomarkers and therapeutic targets. However, the majority of lncRNAs have unknown function and mechanisms of action.

Compounding this problem is that lncRNAs are not evolutionarily constrained at the sequence level and can evolve function rapidly. Indeed, sequence comparison methods (e.g. BLAST) largely fail to identify meaningful similarity between most lncRNAs. Thus, there is an urgent need for new computational tools to identify function in lncRNAs. Historically, evolutionary information has proved extremely useful for this task because mutations disrupting vital functions are removed by natural selection. However, current approaches to identify DNA that is conserved across evolutionary timescales rely on sequence comparison methods that are not suited for lncRNAs. We hypothesize that, for many lncRNAs, it is the complement of protein binding sites rather their order along the molecule that determines function.

We therefore propose a new approach, in which conservation across a phylogenetic tree is quantified at the level of k-mers rather than linear sequence. We will validate our approach by applying it to 100 vertebrate genomes and by comparing our computational results to experimental measurements of RNA structure and protein binding. If successful, our approach will yield a new paradigm for identifying conserved DNA, and will lead to an NIH R01 application.


“A topological model of the cell cycle in KRAS-driven cancers”

  • Co-PI: Jeremy Purvis, Ph.D., Associate Professor, Department of Genetics
  • Co-Pi: Channing Der, Ph.D., Professor, Department of Pharmacology

This team’s research objectives: The goal of this pilot project is to develop a computational model of KRAS-driven cancers that is inspired by concepts in the field of topology. The model will be used to identify specific pathways that enable cells to proliferate when overexpressing a mutant form of the oncogene, KRAS. Mutations in KRAS are the most prevalent in RAS-driven cancers including pancreatic ductal adenocarcinoma, colorectal cancer and lung adenocarcinomas—the three tumor types responsible for the most cancer deaths in the United States. Overexpression of KRAS accelerates the cell cycle, leading to premature DNA replication and the accumulation of DNA damage. This response causes some cells to arrest while others cells escape arrest and continue to proliferate. The field needs accurate models to describe how the cell cycle is altered under KRAS stimulation in order to understand how to terminate cell cycle progression in these tumor-initiating cells. Borrowing concepts from topology, graph theory, and stochastic simulation, we will build a mathematical model of KRAS-driven cancers that explains how the cell cycle changes in response to oncogenic KRAS. We will first build a directed graph of cell state transitions by linking together multi-dimensional measurements of protein activity from asynchronously dividing pancreatic cells. We will link these singe-cell measurements to time-lapse movies to empirically determine the rates of transition between molecular states along the graph. These measurements will allow us to assign transition probabilities between molecular states and derive a cell cycle model that is specific to a particular cell type and oncogenic transformation. We expect this approach to recapitulate the known order of molecular transitions through the cell cycle, but also to reveal new modalities that characterize KRAS-driven cancers. The development of a data-driven, computational model that describes the unique behavior of the cell cycle under oncogenic KRAS will pave the way for other cancer-specific and patient-specific cell cycle models. This is a joint project between the laboratories of Dr. Jeremy Purvis (lead computational PI) and Dr. Channing Der (lead experimental PI) with the ultimate goal of developing preliminary data necessary to compete for a U01 Cancer Systems Biology Award (PAR-19-287).


“A Novel computational pipeline for single cell RNA-seq analysis to reveal the role of type-I IFN in HIV-induced immune dysfunction and viral persistence”

  • Co-PI: Di Wu, Ph.D., Assistant Professor, Adam School of Dentistry
  • Co-PI: Lishan Su, PhD., Professor, Department of Microbiology and Immunology

This team’s research objective: HIV-1 infection continues to be a major global public health issue. Combined antiretroviral therapy (cART) suppresses viral replication and improves survival and quality of life for those HIV-1–infected patients who can both access and tolerate cART. However, cART is not curative and must be continued for life. The HIV-1 reservoir persists indefinitely under suppressive cART, resulting in viral rebound in all HIV-1–infected individuals when cART is discontinued. In addition, despite efficient suppression of HIV-1 replication with cART, low levels of type I interferon (IFN-I) signaling persist in some individuals. Recent breakthrough results show, low levels of IFN-I signaling contribute to HIV-1–associated immune dysfunction and foster HIV-1 persistence in cART-treated hosts. The mechanism(s) that underlie these effects, however, are not clear. We hypothesize that persistent IFN-I signaling leads to dysfunction of multiple subsets of immune cells, which ultimately cause aberrant inflammation and HIV-1 persistence. We propose to develop a novel computational pipeline using existing single cell RNAseq analysis tools and the statistical tools we developed to understand how HIV infection and persistent IFN-I signaling affect immune cell gene expression , cell signaling pathways, cell-cell interactions, cell differentiation and the compositions of immune cell subpopulations. The results from this proposed work will help to find anti IFN-1 realted treatment for HIV-1 persistence.


Thank you to all of the applicants and congratulations to the winners.