This repository hosts the code used to analyze both transcriptomic and morphology imaging data in the paper "" by Lengerich et al, 2022.
R code for transcriptomics can be found in the "transcriptomics" folder as a series of R Markdown files. The output of each R Markdown is used as input to the next R Markdown file. If you wish to regenerate the single-cell analyses from scratch, please download the raw data files from GEO and process them using Cell Ranger as described in the paper. The Cell Ranger outputs are then used in the first R Markdown file in each Study subfolder. If you wnat to skip these early processing steps and regerate the final anlyses directly, we have also uploaded to GEO a counts matrix and cell-level metadata table for each single-cell study. Together, these can be used to build the exact final Seurat objects used to make figures.
R Packages and the versions used are provided in the renv.lock file. This can be used with the renv package to reproduce the environment used in this analysis.
The python code for the morphology analysis is found under the morphology subdirectory organized under the atv_trem2_morpho python package. In brief, to reproduce the analysis, install the
atv_trem2_morpho package into a fresh python 3.10 virtual environment, then run the
script:
python3.10 scripts/plot_seg_stats.py morpho cd74 axl
Which will reproduce the panels in Supplemental Figure 2. See morphology/README.md for
details on installation, command line arguments, and supported OS/Python versions.