Iteratively reweighted least squares (IRLS) procedure in CS-CORE (base R, archived)

This function was originally implemented in 2023 and included in the first release of the CS-CORE R package. It has since been replaced by CSCORE_IRLS for two main reasons:

Usage

CSCORE_IRLS_base(X, seq_depth, post_process = TRUE)

Source

Su, C., Xu, Z., Shan, X., Cai, B., Zhao, H., & Zhang, J. (2023). Cell-type-specific co-expression inference from single cell RNA-sequencing data. Nature Communications. doi: https://doi.org/10.1038/s41467-023-40503-7

Arguments

X: A n by p matrix of UMI counts, where n denotes the number of cells and p denotes the number of genes
seq_depth: A length n vector of sequencing depths
post_process: Whether to process the estimated co-expressions such that the estimates are between -1 and 1. Default to TRUE.

Value

A list of three p by p matrices:

est: co-expression estimates
p_value: p values
test_stat: test statistics

Details

It relies on base R for regression, which is slower and more memory-intensive than the Rcpp-based implementation in CSCORE_IRLS.
It does not support covariate adjustment, a feature supported in CSCORE_IRLS.

Note

This function is retained for reference and backward compatibility, but users are encouraged to use CSCORE_IRLS for new analyses.

Examples

## Toy example:
## run CSCORE on a simulated independent gene pair
cscore_example <- CSCORE_IRLS_base(ind_gene_pair$counts, ind_gene_pair$seq_depths)
#> [1] "IRLS converged after 2 iterations."
#> [1] "0.0000% co-expression estimates were greater than 1 and were set to 1."
#> [1] "0.0000% co-expression estimates were smaller than -1 and were set to -1."

## Estimated co-expression between two genes
cscore_example$est[1,2]
#> [1] 0.007820124
# close to 0: 0.007820124

## p-values
cscore_example$p_value[1,2]
#> [1] 0.961981
# not significant: 0.961981