Robustness and applicability of transcription factor and pathway analysis tools on single-cell RNA-seq data
Background
Many functional analysis tools have been developed to extract functional and mechanistic insight from bulk transcriptome data. With the advent of single-cell RNA sequencing (scRNA-seq), it is in principle possible to do such an analysis for single cells. However, scRNA-seq data has characteristics such as drop-out events and low library sizes. It is thus not clear if functional TF and pathway analysis tools established for bulk sequencing can be applied to scRNA-seq in a meaningful way.
Results
To address this question, we perform benchmark studies on simulated and real scRNA-seq data. We include the bulk-RNA tools PROGENy, GO enrichment, and DoRothEA that estimate pathway and transcription factor (TF) activities, respectively, and compare them against the tools SCENIC/AUCell and metaVIPER, designed for scRNA-seq. For the in silico study, we simulate single cells from TF/pathway perturbation bulk RNA-seq experiments. We complement the simulated data with real scRNA-seq data upon CRISPR-mediated knock-out. Our benchmarks on simulated and real data reveal comparable performance to the original bulk data. Additionally, we show that the TF and pathway activities preserve cell type-specific variability by analyzing a mixture sample sequenced with 13 scRNA-seq protocols. We also provide the benchmark data for further use by the community.
Conclusions
Our analyses suggest that bulk-based functional analysis tools that use manually curated footprint gene sets can be applied to scRNA-seq data, partially outperforming dedicated single-cell tools. Furthermore, we find that the performance of functional analysis tools is more sensitive to the gene sets than to the statistic used.
The datasets supporting the conclusions of this publication are available at Zenodo:
From Zenodo you can download these (zipped) folders:
data- Contains raw data for the analysesoutput- Contains intermediate and final results
Please deposit the unzipped folders in the root directory of this R-project.
Exceptions:
- The raw Human Cell Atlas data (
"data/hca_data/expression_data/sce.all.technologies.RData") are not available on Zenodo. Instead the users can work with the normalized data stored in"output/hca_data/expression/norm.rds". The raw data are accessible on GEO: GSE133549. - Only a small fraction of the raw and normalized expression data of the simulated single cells are avaiable. All data together exceed the limitation of Zenodo's maximal upload size.
Holland CH, Tanevski J, Perales-Patón J, Gleixner J, Kumar MP, Mereu E, Joughin BA, Stegle O, Lauffenburger DA, Heyn H, Szalai B, Saez-Rodriguez, J. "Robustness and applicability of transcription factor and pathway analysis tools on single-cell RNA-seq data." Genome Biology. 2020. DOI: 10.1186/s13059-020-1949-z.
Analysis script available here.
Analysis script available here.
Analysis script available here.
Analysis script available here.
We provide also scripts for plotting individual figures and collages.