Renin cell identity

Here we include a step-by-step breakdown of the analyses to reproduce the data presented in "The reninness score: integrative analysis of multi-omic data to define renin cell identity".

The goal is to identify a unique epigenetic landscape that defines renin cell identity; and develop a computational tool to use that unique epigenetic landscape to identify renin-expressing cells, and quantify the renin program of unknown cell samples.

Overview of the experimental design

Read preprocessing

We used the PEPATAC pipeline to process the raw ATAC-seq reads, including alignment, peak-calling, and quality control. The input files to run PEPATAC for this study are stored in the metadata sub-folder. For more information on how to use PEPATAC, see: http://pepatac.databio.org/

Consensus peak set generation

We used the genomic interval machine learning (geniml) Python package to construct a consensus region set, or the “universe”, using maximum likelihood approach. For more information on how to use genimal, see: https://docs.bedbase.org/geniml/tutorials/create-consensus-peaks/

Differential accessibility analysis and differential accessibile region annotation

All code used for differential accessibility analysis and differential accessibile region annotation are stored in the src sub-folder.

Reninness score calculation and model performace evaluation

All code used for model training and reninness score calulation can be found here). All code used for model performace evaluation are stored in the src sub-folder.