High-throughput screening and classification of chemicals and their effects on neuronal gene expression using RASL-seq

This page contains additional data supporting our manuscript published March 2019 in Scientific Reports

Abstract

We previously used RNA-seq to identify chemicals whose effects on neuronal gene expression mimicked transcriptional signatures of autism, aging, and neurodegeneration. However, this approach was costly and time consuming, which limited our study to testing a single chemical concentration on mixed sex cortical neuron cultures. Here, we adapted a targeted transcriptomic method (RASL-seq, similar to TempO-seq) to interrogate changes in expression of a set of 56 signature genes in response to a library of 350 chemicals and chemical mixtures at four concentrations in male and female mouse neuronal cultures. This enabled us to replicate and expand our previous classifications, and show that transcriptional responses were largely equivalent between sexes. Overall, we found that RASL-seq can be used to accelerate the pace at which chemicals and mixtures that transcriptionally mimic autism and other neuropsychiatric diseases can be identified, and provides a cost-effective way to quantify gene expression with a panel of marker genes.  

Below you’ll find an interactive heatmap that was built using Clustergrammer2. The tool allows for panning and zooming; chemicals, concentrations, sex, and chemical classifications are visible following zoom-in. Given the large size of this dataset, please be patient as it loads and renders. Note: chemical and gene ordering won’t exactly match that of Fig. 2 in the manuscript due to slight differences in clustering algorithm used.