ko, S. | bioRxiv Science

bioRxiv2021

β-Catenin-NFκB-CFTR interactions in cholangiocytes regulate inflammation and fibrosis during ductular reaction

Expansion of biliary epithelial cells (BECs) during ductular reaction (DR) is observed in liver diseases including cystic fibrosis (CF), and associated with inflammation and fibrosis, albeit without complete understanding of underlying mechanism. Using two different genetic knockouts of {beta}-catenin, one with {beta}-catenin loss is hepatocytes and BECs (KO1), and another with loss in only hepatocytes (KO2), we demonstrate disparate long-term repair after an initial injury by 2-week choline-deficient ethionine- supplemented diet. KO2 show gradual liver repopulation with BEC-derived {beta}-catenin- positive hepatocytes, and resolution of injury. KO1 showed persistent loss of {beta}-catenin, NF-{kappa}B activation in BECs, progressive DR and fibrosis, reminiscent of CF histology. We identify interactions of {beta}-catenin, NF{kappa}B and CF transmembranous conductance regulator (CFTR) in BECs. Loss of CFTR or {beta}-catenin led to NF-{kappa}B activation, DR and inflammation. Thus, we report a novel {beta}-catenin-NF{kappa}B-CFTR interactome in BECs, and its disruption may contribute to hepatic pathology of CF.

pathology↗

bioRxiv2021

Computationally scalable regression modeling for large-scale clinical omics data with ParProx

Statistical analysis of ultrahigh-dimensional omics scale data has long depended on univariate hypothesis testing. With growing data features and samples, the obvious next step is to establish multivariable association analysis as a routine method for understanding genotype-phenotype associations. Here we present ParProx, a state-of-the-art implementation to optimize overlapping group lasso regression models for time-to-event and classification analysis, guided by biological priors through coordinated variable selection. ParProx not only enables model fitting for ultrahigh-dimensional data within the architecture for parallel or distributed computing, but also allows users to obtain interpretable regression models consistent with known biological relationships among the independent variables, a feature long neglected in statistical modeling of omics data. We demonstrate ParProx using three different omics data sets of moderate to large numbers of variables, where we use genomic regions and pathways to arrive at sparse regression models comprised of biologically related independent variables. ParProx is naturally applicable to a wide range of studies using ultrahigh-dimensional omics data, ranging from genome-wide association analysis to single cell sequencing studies where multivariable modeling is computationally intractable.

bioinformatics↗