/scRNA_disease

Code for our paper: https://www.biorxiv.org/content/10.1101/528463v1

Primary LanguageHTML

Integrating GWAS with single cell RNA-seq

Last update: 24.4.2020

The paper is currently on bioRxiv and can be accessed here.

Requirements

  1. R libraries
install.packages("tidyverse")
install.packages("AnnotationDbi")
install.packages("org.Hs.eg.db")

  1. MAGMA and its auxiliary files

  2. Partitioned LDSC and its auxiliary files

Specificity files

By following step 1) on your GWAS, you should then be able to run quickly the MAGMA association code and the LDSC code.

Steps

The code to reproduce our results is located in this repository.

The following links show the essential steps:

  1. Get GWAS summary statistics in the right format for MAGMA and LDSC

  2. Get MAGMA and LDSC input for the Zeisel et al. data set

  3. Get MAGMA and LDSC input for the GTEx et al. data set

  4. Get MAGMA and LDSC input for the Skene et al. data set

  5. Get MAGMA and LDSC input for the Habib et al. data set

  6. Get MAGMA and LDSC input for the Saunders et al. data set

  7. Get MAGMA and LDSC input for the Lake et al. data set

Run MAGMA

Once the GWAS sumstats are ready and the specificity files are ready, you can use the following code to test for associations using MAGMA.

If you want to run your GWAS with our specificity files, you just need to get the 'top10.txt' files in the different MAGMA folders.

Run LDSC

Once the GWAS sumstats are ready and the specificity files are ready, you can use the following code to test for associations using LDSC.

Alternatively, you could use a nextflow pipeline located here

The code to look for heritability enrichment of the top 10% most specific genes was made to be run in parallele on a SLURM cluster. It would need to be adapted if you want to run it locally.

Expression Weighted Celltype Enrichment

The paper describing EWCE is accessible here.

EWCE code is accessible here.