This repository contains R scripts to accompany: Ellender TJ, Avery SV, Mahfooz K, et al. Embryonic progenitor pools generate diversity in fine-scale excitatory cortical subnetworks. Nat Commun. 2019;10(1):5224. Published 2019 Nov 19. doi:10.1038/s41467-019-13206-1
- Install https://github.com/cgat-developers/cgat-core and https://github.com/cgat-developers/cgat-apps
- Install https://github.com/AllenInstitute/scrattch.hicat
- Download the project057 github repository
- Run
`setup.py develop`in the project057 directory
The scripts require a `.tsv` counts table, generated e.g. by featurecounts as input.
They also require an annotation table for the columns.
Run the cgat-singlecell --help command to see what scripts are available and how to use them.
For example, to run filter data obtained from featurecounts
`cgat-singlecell filter --counts-filename=featurecounts.tsv --phenotypes-filename=phenodata.tsv --factor=group,mouse_id,collection_date,slice_depth,slice_number,pipette_visual,timepoint > filtered_counts.tsv`
To run the normalisation script and map the data onto a reference dataset (e.g. Allen) use:
`cgat-singlecell normalisation --rds-filename sce_filtered_hicat.rds --ERCC ERCC.tsv --allen-design mouse_VISp_2018-06-14_samples-columns.csv --allen-datamatrix mouse_VISp_2018-06-14_exon-matrix.csv --allen-rowdata mouse_VISp_2018-06-14_genes-rows.csv --allen-filter "L4 IT" --norm scran --colours red,green --allen-colours black --perplexity 5 > pipeline.log`
For the publication the following sequence was used on each experimental plate separately.
- Filtering of datase using
`cgat-singlecell filtering` - Run scrattch.hicat using wrapper script
`cgat-singlecell hicat` - Normalisation and mapping to reference dataset using
`cgat-singlecell normalisation` - Differential expression analysis using
`cgat-singlecell sc-diffexpression`