1 Abstract

This year has seen the number of dedicated tools for the analysis of scRNA-seq data reach over 250, with over half implemented in R. Most of these tools address only parts of an scRNA-seq data analysis workflow, like alignment, normalization or clustering. Even when tools cover multiple parts of the analysis workflow, their selection of methods may not present the best solution according to ongoing benchmarking efforts in the field or may be unsuitable for more complex data. This forces users not just to switch between tools, but also between object classes and in some cases programming languages.

Here we want to present our pipeline for the analysis of single cell nuclei RNA-sequencing (sc/n-RNA-seq) of surgically resected brain tissue from four patients with brain malformations. For each patient we have data from both normal as well as dysplastic tissue. While this powerful experimental design allows us to reveal the molecular characteristics of brain malformations, such as dysplastic cell types, it requires a non-standard pipeline of many tools. We will report on the challenges we encountered due to this increased modularity. We will also report on data integration efforts across individuals as well as interactive tools developed during this project to improve communication with biologists and clinicians.