fastCorrDiff-package: Fast Differential Correlation Matrix Screening
In fastCorrDiff: Fast Differential Correlation Matrix Screening

Description Details Author(s) References Examples

Fast spectral algorithms for differential analysis on large-scale correlation matrices of Li et. al. (2021) <arXiv:2111.03721> are implemented. The methods can identify a group of genes exhibiting differential correlation patterns between two groups. For high-dimensional matrices, compressed spectral screening methods are also provided based on random subsampling. The work to build this package is partially supported by the NSF grant DMS-2015298.

The DESCRIPTION file: This package was not yet installed at build time.
Index: This package was not yet installed at build time.
The package for fast differential correlation/covariance analysis based on spectral methods. It includes the spectral screening and compressed spectral screening methods proposed in Li et al (2021) <arXiv:2111.03721> to identify a group of variables exhibiting differential correlation pattern between two settings.The methods are designed for a great scalability when the number of variables is large, such that computing the full correlation matrices is not feasible.

Tianxi Li and Xiwei Tang

Maintainer: Tianxi Li <tianxili@virginia.edu>

Tianxi Li, Xiwei Tang, and Ajay Chatrath. Compressed spectral screening for large-scale differential correlation analysis with application in selecting Glioblastoma gene modules. arXiv preprint arXiv:2111.03721, 2021.

T. T. Cai, T. Liang, A. Rakhlin, et al. Computational and statistical boundaries for submatrix localization in a large noisy matrix. The Annals of Statistics, 45(4):1403-1430, 2017.

X1 <- matrix(rnorm(200),10,20)
C1 <- cor(X1)
X2 <- matrix(rnorm(200),10,20)
C2 <- cor(X2)

D <- C1-C2

fit <- SS(D, 3, K.seq = FALSE, sv = FALSE)

fit$score