igm-team / subrvis

MIT License
4 stars 3 forks source link

SubRVIS

This repository contains the software used in the subRVIS publication (http://bit.ly/2vY1KP5).

Citation

Gussow AB, Petrovski S, Wang Q, Allen AS, Goldstein DB. The intolerance to functional genetic variation of protein domains predicts the localization of pathogenic mutations within genes. Genome Biology. 2016; 17(1):9.

Regional Score Gene-Specific Prediction Test Model

To run the single gene assessment used in the manuscript, the usage is:

Rscript src/single_gene_tests.R -s <scores> -c <counts> --perms <number of permutations> -n <minimum pathogenic variants> -m <minimum regions> -o <out file path>

For example:

Rscript src/single_gene_tests.R -s data/domains.rvis -c data/domains_counts.txt --perms 20000 -n 1 -m 2 -o subrvis_single_gene_test_results.txt

Regional Score Genome-Wide Prediction Test Model

To run the genome-wide assessment used in the manuscript, the usage is:

Rscript src/genome_wide_test.R --table <scores table> --genes <gene list> --covars <predictor names> --out <out file path>

For example:

Rscript src/genome_wide_test.R --table data/domains_table.txt --genes data/genes.txt --covars subRVIS --out subrvis_genome_wide_test_results.txt

subRVIS Variant Plotting

The variant_plotting directory contains the subRVIS website tool (www.subrvis.org), which is based on the Shiny framework (http://shiny.rstudio.com).