Single-cell alternative polyadenylation analysis delineates GABAergic neuron types.

BACKGROUND: Alternative polyadenylation (APA) is emerging as an important mechanism in the post-transcriptional regulation of gene expression across eukaryotic species. Recent studies have shown that APA plays key roles in biological processes, such as cell proliferation and differentiation. Single-cell RNA-seq technologies are widely used in gene expression heterogeneity studies; however, systematic studies of APA at the single-cell level are still lacking. RESULTS: Here, we described a novel computational framework, SAPAS, that utilizes 3'-tag-based scRNA-seq data to identify novel poly(A) sites and quantify APA at the single-cell level. Applying SAPAS to the scRNA-seq data of phenotype characterized GABAergic interneurons, we identified cell type-specific APA events for different GABAergic neuron types. Genes with cell type-specific APA events are enriched for synaptic architecture and communications. In further, we observed a strong enrichment of heritability for several psychiatric disorders and brain traits in altered 3' UTRs and coding sequences of cell type-specific APA events. Finally, by exploring the modalities of APA, we discovered that the bimodal APA pattern of Pak3 could classify chandelier cells into different subpopulations that are from different laminar positions. CONCLUSIONS: We established a method to characterize APA at the single-cell level. When applied to a scRNA-seq dataset of GABAergic interneurons, the single-cell APA analysis not only identified cell type-specific APA events but also revealed that the modality of APA could classify cell subpopulations. Thus, SAPAS will expand our understanding of cellular heterogeneity.

Computational modeling and analysis of the morphogenetic domain signaling networks regulating C. elegans embryogenesis.

Caenorhabditis elegans, often referred to as the 'roundworm', provides a powerful model for studying cell autonomous and cell-cell interactions through the direct observation of embryonic development in vivo. By leveraging the precisely mapped cell lineage at single cell resolution, we are able to study at a systems level how early embryonic cells communicate across morphogenetic domains for the coordinated processes of gene expressions and collective cellular behaviors that regulate tissue morphogenesis. In this study, we developed a computational framework for the exploration of the morphogenetic domain cell signaling networks that may regulate C. elegans gastrulation and embryonic organogenesis. We demonstrated its utility by producing the following results, i) established a virtual reference model of developing C. elegans embryos through the spatiotemporal alignment of individual embryo cell nuclear imaging samples; ii) integrated the single cell spatiotemporal gene expression profile with the established virtual embryo model by data pooling; iii) trained a Machine Learning model (Random Forest Regression), which predicts accurately the spatial positions of the cells given their gene expression profiles for a given developmental time (e.g. total cell number of the embryo); iv) enabled virtual 4-dimensional tomographic graphical modeling of single cell data; v) inferred the biology signaling pathways that act in each of morphogenetic domains by meta-data analysis. It is intriguing that the morphogenetic domain cell signaling network seems to involve some crosstalk of multiple biology signaling pathways during the formation of tissue boundary pattern. Lastly, we developed the Software tool 'Embryo aligner version 1.0' and provided it as an Open Source program to the research community for virtual embryo modeling, and phenotype perturbation analyses (https://github.com/csniuben/embryo_aligner/wiki and https://bioinfo89.github.io/C.elegansEmbryonicOrganogenesisweb/).

Comparing therapeutic modulators of the SOD1 G93A Amyotrophic Lateral Sclerosis mouse pathophysiology.

Introduction: Amyotrophic Lateral Sclerosis (ALS) is a paralyzing, multifactorial neurodegenerative disease with limited therapeutics and no known cure. The study goal was to determine which pathophysiological treatment targets appear most beneficial. Methods: A big data approach was used to analyze high copy SOD1 G93A experimental data. The secondary data set comprised 227 published studies and 4,296 data points. Treatments were classified by pathophysiological target: apoptosis, axonal transport, cellular chemistry, energetics, neuron excitability, inflammation, oxidative stress, proteomics, or systemic function. Outcome assessment modalities included onset delay, health status (rotarod performance, body weight, grip strength), and survival duration. Pairwise statistical analysis (two-tailed t-test with Bonferroni correction) of normalized fold change (treatment/control) assessed significant differences in treatment efficacy. Cohen's d quantified pathophysiological treatment category effect size compared to "all" (e.g., all pathophysiological treatment categories combined). Results: Inflammation treatments were best at delaying onset (d = 0.42, p > 0.05). Oxidative stress treatments were significantly better for prolonging survival duration (d = 0.18, p < 0.05). Excitability treatments were significantly better for prolonging overall health status (d = 0.22, p < 0.05). However, the absolute best pathophysiological treatment category for prolonging health status varied with disease progression: oxidative stress was best for pre-onset health (d = 0.18, p > 0.05); excitability was best for prolonging function near onset (d = 0.34, p < 0.05); inflammation was best for prolonging post-onset function (d = 0.24, p > 0.05); and apoptosis was best for prolonging end-stage function (d = 0.49, p > 0.05). Finally, combination treatments simultaneously targeting multiple pathophysiological categories (e.g., polytherapy) performed significantly (p < 0.05) better than monotherapies at end-stage. Discussion: In summary, the most effective pathophysiological treatments change as function of assessment modality and disease progression. Shifting pathophysiological treatment category efficacy with disease progression supports the homeostatic instability theory of ALS disease progression.