BrainTrawler Lite: Navigating through a multi-scale multi-modal gene transcriptomics data resource through a lightweight user interface

Joint exploration of behavior, genes and brain structure holds the promise to provide a better understanding of neural circuits than unimodal data analyses. Several consortia provide extensive data resources of different modalities. However, analyses of heteromodal data require the expertise of computer scientists for fusing the different resources into a common space. To tackle this limitation, we extended BrainTrawler, our web-based visual analytics frame work for exploration of neurobiological data of human and mice, with a selection of publicly available gene expression and connectivity datasets. RNA sequencing datasets were combined to cover the full brain in both species, partly including cell type information. For an easy navigation through this unique data collection, we additionally created BrainTrawler Lite, a lightweight user interface. It provides a visual overview of gene expression information available across resources in the form of an interactive heat map and tools for deeper exploration of available data (see Figure). Overall, our extension of BrainTrawler provides neuroscientists with simple access to a unique collection of connectivity and gene expression data up to cell-type resolution that can be jointly explored for hypothesis generation on neural circuits. BrainTrawler is publically available at braintrawler.vrvis.at.Figure: BrainTrawler Lite interface a) A heat map shows the information coverage across datasets (rows) and brain regions (columns). b) Expression is shown across selected data sets and brain regions for a set of five genes using small multiples. c) Parallel coordinates visualizing gene expression of all genes in selected datasets. Brushing of coordinates allows gene filtering.