sslmm: Computing Summary-level Data from Individual-level Data

In FLASHMM: Fast and Scalable Single Cell Differential Expression Analysis using Mixed-Effects Models

Computing Summary-level Data from Individual-level Data

Description

sslmm can be used to compute the summary statistics (summary-level data) for lmm function, defined as

• XX = t(X)%*%X

• XY = t(X)%*%t(Y)

• ZX = t(Z)%*%X

• ZY = t(Z)%*%t(Y)

• ZZ = t(Z)%*%Z

• Ynorm = rowSums(Y*Y)

• n = nrow(X)

Usage

sslmm(X, Y, Z, nBlocks = ceiling((ncol(Y) * 1e-08) * nrow(Y)))

Arguments

X	A design matrix for fixed effects, with rows corresponding to the columns of Y.
Y	A features-by-samples matrix of responses (genes-by-cells matrix of gene expressions for scRNA-seq).
Z	A design matrix for random effects, with rows corresponding to the columns of Y.
nBlocks	Number of the blocks, which a big data is subdivided into, used for reducing the storage in computing the summary statistics that are computed from a block of data. The default value may not be adequate. If encountering the error: vector memory limit reached, you should increase the nBlocks value to avoid the issue.

Value

A list of summary statistics: XX, XY, ZX, ZY, ZZ, Ynorm and n.

Examples

n <- 1e3
set.seed(2024)
p <- 2
X <- matrix(rnorm(p*n), n, p)
colnames(X) <- paste0("X", 1:ncol(X))
m <- 3
Y <- matrix(rnorm(m*n), m, n)
rownames(Y) <- paste0("Y", 1:nrow(Y))
q <- 4
Z <- gl(q, n/q, labels = letters[1:q])
Z <- model.matrix(~ 0 + Z)
sslmm(X, Y, Z)

s1 <- sslmm(X, Y, Z, nBlocks = 1)
s2 <- sslmm(X, Y, Z, nBlocks = 2)
s3 <- sslmm(X, Y, Z, nBlocks = 3)

identical(s1, s2)
identical(s2, s3)

FLASHMM documentation built on Aug. 21, 2025, 5:49 p.m.