open_bicor {TRN} R Documentation

re-open the persisted bicor correlation store that was written by `write_bicor`

Description

Usage

open_bicor(repo,
    cache = 2048);

Arguments

repo

the data repository directory that was passed to write_bicor (reads {repo}/bicor.dat, or {repo}/pearson.dat when the store was written with corKind = "pearson", and {repo}/index.dat). [as string]

cache

the hot-row LRU cache capacity in rows. Each cached row of a N = 50,000 matrix costs about 200 KB of memory. The default value 2048 means at most about 400 MB of the hot row cache. [as integer]

env

the R# runtime environment object. [as Environment]

Details

the returned store supports fast point queries (gene1, gene2 -> correlation and p-value), neighborhood queries (gene + threshold -> all correlated genes) and full-library streaming edge filters, all without loading the matrix into memory.

Authors

cytoscape

Value

an opened CorrelationMatrixStore object. Pass it to build_grn via the bicor parameter to skip the expensive correlation matrix re-computation.

clr value class

Examples

 let bicor = open_bicor("Z:/hsa_mat");

[Package TRN version 1.0.0.0 Index]