The PLRC Grant Review committee has selected the awardees for the 2023-2024 Pilot & Feasibility Program Grants. Our thanks to all who submitted proposals.
CONGRATULATIONS to the following awardees of the 2023 PLRC Pilot and Feasibility Program Grants!
Deep learning-based prediction of drug repurposing and genetic vulnerabilities in hepatocellular carcinoma.
Hepatoceullar carcinoma (HCC) is the major type of liver cancer, which is the only cancer type with increasing incidence and mortality rates in men and women in the United States1, showing the need for better therapy. Pharmacogenomics could be a key to cures by studying how integrated genomic characteristics influence cancer response to small molecules (drug sensitivity) and genetic perturbations (gene knockouts to study genetic dependencies). At the pan-cancer level, global consortia have molecularly profiled a huge number of tumors2 and systematically screened for small molecules and genetic perturbations in a panel of pan-cancer cell lines that broadly represent the heterogeneity of tumors3-6. The emerging data resources have enabled deep learning, a powerful branch of machine learning, to discover intricate pharmacogenomics associations7. We recently developed two pioneering deep learning models that accurately predicted the response of pan-cancer cell lines to perturbations by integrative genomics profiles8, 9. However, because of the scarcity of HCC samples in the pancancer data (<2%), sophisticated computational models are required, but lacking, to efficiently capture genomic features and vulnerabilities specifically for this challenging malignancy.
The proposed study will address this research gap by developing novel deep learning models that are optimized on extensively collected HCC data. Our long-term goal is to increase the understanding of the pharmacogenomics of HCC and facilitate the development of small molecules and novel targets for future therapies. We hypothesize that genomics profiles integrated by deep learning predict responses of HCC cells to drug and genetic perturbations.
Understanding Mechanisms of Progressive Familial Intrahepatic Cholestasis Type 1 (PFIC1) Using Patient- Specific Inducible Pluripotent Stem Cells
Progressive familial intrahepatic cholestasis (PFIC) refers to a heterogeneous group of disorders that are linked by the inability to appropriately form and excrete bile from hepatocytes, resulting in a hepatocellular form of cholestasis. While the diagnosis of such disorders had historically been based on pattern recognition of unremitting cholestasis without other identified molecular or anatomic cause, recent scientific advancements have enabled identification of multiple sub-types defined by the dysfunction of specific responsible proteins (1). The first described PFIC, type 1 (PFIC1), arises from biallelic variants in the ATP8B1 gene that annul activity of the FIC1 protein located at the canalicular membrane of hepatocytes. Unfortunately, scientific discovery and therapeutic advancement for PFIC1 have been hampered by animal models which fail to recapitulate disease pathobiology and a rarity that prohibits definitive trials in human subjects. This reality has led to on-going investigations and controversy regarding the exact molecular pathophysiology driving development of FIC1- associated liver disease. Having a functional platform for mechanistic investigations in PFIC1 would provide a critical unmet need for scientific exploration in this (and other) cholestatic condition. Recently, the ability to derive patient-specific induced pluripotent stem cell (iPSCs) has demonstrated novel opportunities to model hepatobiliary disorders, gain mechanistic insights, and explore cell-based therapies. Briefly, human fibroblasts or lymphocytes are isolated and reprogrammed into hepatocyte-like cells (HLC) which bear significant similarities to their primary counterparts in terms of transcriptional profile and functional properties, but retain the unique, patient specific genetic defects (2, 3). These patient-derived iPSC lines enable studying individual genotypes under highly controlled conditions and allow linking the observed defects directly to disease-causing genetic alterations (4). While various liver-specific disease models have been established (2, 5), there remains a relative paucity of data as to the potential of iPSCs in the study of PFIC1 (6, 7). We postulate that human iPSCs derived from reprogrammed somatic cells of patients with PFIC1 can be used to further explore and expand the knowledge base of this rare disease. Thus, in the present study, we will examine the overarching hypothesis that PFIC1-specific HLCs accurately recapitulate characteristics of primary hepatocytes and can further enhance understanding of mechanisms driving development of the PFIC1 phenotype.