tests/testthat/test_testDiffExp.R
In MikeDMorgan/miloR: Differential neighbourhood abundance testing on a graph
context("Testing post-hoc DGE function")
library(miloR)
### Set up a mock data set using simulated data
library(SingleCellExperiment)
library(scran)
library(scater)
library(irlba)
library(MASS)
library(mvtnorm)
set.seed(42)
r.n <- 1000
n.dim <- 50
block1.cells <- 500
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block1.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block1.eigens <- block1.eigens[order(block1.eigens)]
block1.p <- qr.Q(qr(matrix(rnorm(block1.cells^2, mean=4, sd=0.01), block1.cells)))
block1.sigma <- crossprod(block1.p, block1.p*block1.eigens)
block1.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block1.cells, mean=2, sd=0.01), sigma=block1.sigma))
# need to through in random 0's for realism
block1.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block1.gex), byrow=TRUE)
block1.gex[block1.drop] <- 0
block2.cells <- 500
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block2.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block2.eigens <- block2.eigens[order(block2.eigens)]
block2.p <- qr.Q(qr(matrix(rnorm(block2.cells^2, mean=3, sd=0.01), block2.cells)))
block2.sigma <- crossprod(block2.p, block2.p*block2.eigens)
block2.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block2.cells, mean=4, sd=0.01), sigma=block2.sigma))
# need to through in random 0's for realism
block2.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block2.gex), byrow=TRUE)
block2.gex[block2.drop] <- 0
block3.cells <- 650
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block3.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block3.eigens <- block3.eigens[order(block3.eigens)]
block3.p <- qr.Q(qr(matrix(rnorm(block3.cells^2, mean=4.5, sd=0.01), block3.cells)))
block3.sigma <- crossprod(block3.p, block3.p*block3.eigens)
block3.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block3.cells, mean=5, sd=0.01), sigma=block3.sigma))
# need to through in random 0's for realism
block3.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block3.gex), byrow=TRUE)
block3.gex[block3.drop] <- 0
block4.cells <- 250
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block4.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block4.eigens <- block4.eigens[order(block4.eigens)]
block4.p <- qr.Q(qr(matrix(rnorm(block4.cells^2, mean=4.75, sd=0.01), block4.cells)))
block4.sigma <- crossprod(block4.p, block4.p*block4.eigens)
block4.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block4.cells, mean=5.28, sd=0.01), sigma=block4.sigma))
# need to through in random 0's for realism
block4.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block4.gex), byrow=TRUE)
block4.gex <- sapply(1:ncol(block4.gex), FUN=function(X){
ifelse(block4.drop[, X], 0, block4.gex[, X][!block4.drop[, X]])
})
sim1.gex <- do.call(cbind, list("b1"=block1.gex, "b2"=block2.gex, "b3"=block3.gex, "b4"=block4.gex))
colnames(sim1.gex) <- paste0("Cell", 1:ncol(sim1.gex))
rownames(sim1.gex) <- paste0("Gene", 1:nrow(sim1.gex))
sim1.pca <- prcomp_irlba(t(sim1.gex), n=50, scale.=TRUE, center=TRUE)
set.seed(42)
block1.cond <- rep("A", block1.cells)
block1.a <- sample(1:block1.cells, size=floor(block1.cells*0.9))
block1.b <- setdiff(1:block1.cells, block1.a)
block1.cond[block1.b] <- "B"
block2.cond <- rep("A", block2.cells)
block2.a <- sample(1:block2.cells, size=floor(block2.cells*0.5))
block2.b <- setdiff(1:block2.cells, block2.a)
block2.cond[block2.b] <- "B"
block3.cond <- rep("A", block3.cells)
block3.a <- sample(1:block3.cells, size=floor(block3.cells*0.5))
block3.b <- setdiff(1:block3.cells, block3.a)
block3.cond[block3.b] <- "B"
block4.cond <- rep("A", block4.cells)
block4.a <- sample(1:block4.cells, size=floor(block4.cells*0.05))
block4.b <- setdiff(1:block4.cells, block4.a)
block4.cond[block4.b] <- "B"
meta.df <- data.frame("Block"=c(rep("B1", block1.cells), rep("B2", block2.cells), rep("B3", block3.cells), rep("B4", block4.cells)),
"Condition"=c(block1.cond, block2.cond, block3.cond, block4.cond),
"Replicate"=c(rep("R1", floor(block1.cells*0.33)), rep("R2", floor(block1.cells*0.33)),
rep("R3", block1.cells-(2*floor(block1.cells*0.33))),
rep("R1", floor(block2.cells*0.33)), rep("R2", floor(block2.cells*0.33)),
rep("R3", block2.cells-(2*floor(block2.cells*0.33))),
rep("R1", floor(block3.cells*0.33)), rep("R2", floor(block3.cells*0.33)),
rep("R3", block3.cells-(2*floor(block3.cells*0.33))),
rep("R1", floor(block4.cells*0.33)), rep("R2", floor(block4.cells*0.33)),
rep("R3", block4.cells-(2*floor(block4.cells*0.33)))))
colnames(meta.df) <- c("Block", "Condition", "Replicate")
# define a "sample" as teh combination of condition and replicate
meta.df$Sample <- paste(meta.df$Condition, meta.df$Replicate, sep="_")
meta.df$Vertex <- c(1:nrow(meta.df))
rownames(meta.df) <- colnames(sim1.gex)
sim1.sce <- SingleCellExperiment(assays=list(logcounts=sim1.gex),
reducedDims=list("PCA"=sim1.pca$x))
sim1.mylo <- Milo(sim1.sce)
# build a graph - this can take a while for large graphs - will need to play
# around with the parallelisation options
sim1.mylo <- buildGraph(sim1.mylo, k=30, d=10)
# define neighbourhoods - this is slow for large data sets
# how can this be sped up? There are probably some parallelisable steps
sim1.mylo <- makeNhoods(sim1.mylo, k=30, prop=0.2, refined=TRUE,
d=10,
reduced_dims="PCA")
sim1.mylo <- calcNhoodDistance(sim1.mylo, d=10)
sim1.mylo <- buildNhoodGraph(sim1.mylo)
sim1.meta <- data.frame("Condition"=c(rep("A", 3), rep("B", 3)),
"Replicate"=rep(c("R1", "R2", "R3"), 2))
sim1.meta$Sample <- paste(sim1.meta$Condition, sim1.meta$Replicate, sep="_")
rownames(sim1.meta) <- sim1.meta$Sample
sim1.mylo <- countCells(sim1.mylo, samples="Sample", meta.data=meta.df)
sim1.res <- testNhoods(sim1.mylo, design=~Condition, fdr.weighting="k-distance",
design.df=sim1.meta[colnames(nhoodCounts(sim1.mylo)), ])
sim1.res <- groupNhoods(sim1.mylo, sim1.res)
test_that("Incorrect input gives the expected errors", {
expect_error(testDiffExp(matrix(0L, nrow=nrow(sim1.mylo), ncol=ncol(sim1.mylo))),
"Unrecognised input type")
expect_error(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition, assay="junk"),
"Unrecognised assay slot")
expect_error(suppressWarnings(testDiffExp(sim1.mylo, na.res, meta.data=meta.df,
design=~Condition, na.function="junk")),
"NA function junk not recognised")
# pass a model matrix with wrong dimensions
fake.model <- model.matrix(~Condition, data=meta.df)
fake.model <- fake.model[-c(1:10), ]
expect_error(suppressWarnings(testDiffExp(sim1.mylo, sim1.res,
meta.data=meta.df,
design=fake.model)),
"Cannot subset model matrix, subsetting vector is wrong length")
})
test_that("Less than optimal input gives the expected warnings", {
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition, na.function=NULL,
)),
"NULL passed to na.function, using na.pass")
na.res <- sim1.res
na.res$SpatialFDR[1:10] <- NA
expect_warning(suppressMessages(testDiffExp(sim1.mylo, na.res, meta.data=meta.df,
design=~Condition)),
"NA values found in SpatialFDR vector")
fake.meta <- meta.df
rownames(fake.meta) <- paste0("X", c(1:nrow(meta.df)))
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=fake.meta,
design=~Condition)),
"Column names of x are not the same as meta-data rownames")
fake.model <- model.matrix(~Condition, data=meta.df)
rownames(fake.model) <- paste0("X", c(1:nrow(fake.model)))
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=fake.model)),
"Design matrix and meta-data dimnames are not the same")
# add simulated CPMs to the data
# skip this for now until I can properly debug the issue with a working
# copy of R4.3
# ux.1 <- matrix(rpois((nrow(meta.df)*r.n)/2, 5), ncol=nrow(meta.df))
# ux.2 <- matrix(rpois((nrow(meta.df)*r.n)/2, 4), ncol=nrow(meta.df))
# ux <- rbind(ux.1, ux.2)
# ux.cpm <- apply(ux, 2, function(X) log(X/(sum(X+1)/(1e6+1))))
# colnames(ux.cpm) <- colnames(sim1.mylo)
# rownames(ux.cpm) <- rownames(sim1.mylo)
#
# SingleCellExperiment::cpm(sim1.mylo) <- ux.cpm
# expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
# design=~Condition, assay="cpm")),
# "Assay type is not counts or logcounts")
})
test_that("Output is correct type", {
full.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition))
expect_identical(class(full.out) , "list")
})
test_that("Contrasts can be passed without error" , {
set.seed(42)
block1.cond <- rep("A", block1.cells)
block1.a <- sample(1:block1.cells, size=floor(block1.cells*0.05))
block1.b <- setdiff(1:block1.cells, block1.a)
block1.c <- sample(1:block1.cells, size=floor(block1.cells*0.9))
block1.cond[block1.b] <- "B"
block1.cond[block1.c] <- "C"
block2.cond <- rep("A", block2.cells)
block2.a <- sample(1:block2.cells, size=floor(block2.cells*0.3))
block2.b <- setdiff(1:block2.cells, block2.a)
block2.c <- sample(1:block2.cells, size=floor(block2.cells*0.3))
block2.cond[block2.b] <- "B"
block2.cond[block2.c] <- "C"
block3.cond <- rep("A", block3.cells)
block3.a <- sample(1:block3.cells, size=floor(block3.cells*0.3))
block3.b <- setdiff(1:block3.cells, block3.a)
block3.c <- setdiff(1:block3.cells, sample(1:block3.cells, size=floor(block3.cells*0.3)))
block3.cond[block3.b] <- "B"
block3.cond[block3.c] <- "C"
meta.df <- data.frame("Block"=c(rep("B1", block1.cells), rep("B2", block2.cells), rep("B3", block3.cells)),
"Condition"=c(block1.cond, block2.cond, block3.cond),
"Replicate"=c(rep("R1", floor(block1.cells*0.33)), rep("R2", floor(block1.cells*0.33)),
rep("R3", block1.cells-(2*floor(block1.cells*0.33))),
rep("R1", floor(block2.cells*0.33)), rep("R2", floor(block2.cells*0.33)),
rep("R3", block2.cells-(2*floor(block2.cells*0.33))),
rep("R1", floor(block3.cells*0.33)), rep("R2", floor(block3.cells*0.33)),
rep("R3", block3.cells-(2*floor(block3.cells*0.33)))))
colnames(meta.df) <- c("Block", "Condition", "Replicate")
# define a "sample" as teh combination of condition and replicate
meta.df$Sample <- paste(meta.df$Condition, meta.df$Replicate, sep="_")
meta.df$Vertex <- c(1:nrow(meta.df))
#meta.df$Condition <- ordered(meta.df$Condition, levels=c("A", "B", "C"))
rownames(meta.df) <- colnames(sim1.gex[, c(1:nrow(meta.df))])
blockC.sce <- SingleCellExperiment(assays=list(logcounts=sim1.gex[, c(1:nrow(meta.df))]),
reducedDims=list("PCA"=sim1.pca$x[c(1:nrow(meta.df)), ]))
blockC.mylo <- Milo(blockC.sce)
# build a graph - this can take a while for large graphs - will need to play
# around with the parallelisation options
blockC.mylo <- buildGraph(blockC.mylo, k=30, d=10)
# define neighbourhoods - this is slow for large data sets
# how can this be sped up? There are probably some parallelisable steps
blockC.mylo <- makeNhoods(blockC.mylo, k=30, prop=0.1, refined=TRUE,
d=10, reduced_dims="PCA")
blockC.mylo <- calcNhoodDistance(blockC.mylo, d=10)
blockC.mylo <- buildNhoodGraph(blockC.mylo)
blockC.meta <- data.frame("Condition"=c(rep("A", 3), rep("B", 3), rep("C", 3)),
"Replicate"=rep(c("R1", "R2", "R3"), 3))
blockC.meta$Sample <- paste(blockC.meta$Condition, blockC.meta$Replicate, sep="_")
rownames(blockC.meta) <- blockC.meta$Sample
blockC.mylo <- countCells(blockC.mylo, samples="Sample", meta.data=meta.df)
blockC.res <- testNhoods(blockC.mylo, design=~0 + Condition, fdr.weighting="k-distance",
design.df=blockC.meta[colnames(nhoodCounts(blockC.mylo)), ])
blockC.res <- groupNhoods(blockC.mylo, blockC.res)
# works with estimable contrast levels
expect_error(do.call(rbind.data.frame,
suppressWarnings(testDiffExp(blockC.mylo, blockC.res, meta.data=meta.df,
model.contrasts=c("ConditionC - ConditionB"),
design=~0 + Condition,
gene.offset=FALSE))),
NA)
})
test_that("Row subsetting returns expected number of results", {
full.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~0 + Condition
))
full.lengths <- lapply(full.out, nrow)
subset.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
subset.row=sample(1:nrow(sim1.mylo), size=100),
design=~0 + Condition))
subset.lengths <- lapply(subset.out, nrow)
expect_true(all(unlist(full.lengths) > unlist(subset.lengths)))
})
MikeDMorgan/miloR documentation built on Oct. 19, 2024, 8:39 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context("Testing post-hoc DGE function")
library(miloR)
### Set up a mock data set using simulated data
library(SingleCellExperiment)
library(scran)
library(scater)
library(irlba)
library(MASS)
library(mvtnorm)
set.seed(42)
r.n <- 1000
n.dim <- 50
block1.cells <- 500
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block1.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block1.eigens <- block1.eigens[order(block1.eigens)]
block1.p <- qr.Q(qr(matrix(rnorm(block1.cells^2, mean=4, sd=0.01), block1.cells)))
block1.sigma <- crossprod(block1.p, block1.p*block1.eigens)
block1.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block1.cells, mean=2, sd=0.01), sigma=block1.sigma))
# need to through in random 0's for realism
block1.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block1.gex), byrow=TRUE)
block1.gex[block1.drop] <- 0
block2.cells <- 500
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block2.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block2.eigens <- block2.eigens[order(block2.eigens)]
block2.p <- qr.Q(qr(matrix(rnorm(block2.cells^2, mean=3, sd=0.01), block2.cells)))
block2.sigma <- crossprod(block2.p, block2.p*block2.eigens)
block2.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block2.cells, mean=4, sd=0.01), sigma=block2.sigma))
# need to through in random 0's for realism
block2.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block2.gex), byrow=TRUE)
block2.gex[block2.drop] <- 0
block3.cells <- 650
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block3.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block3.eigens <- block3.eigens[order(block3.eigens)]
block3.p <- qr.Q(qr(matrix(rnorm(block3.cells^2, mean=4.5, sd=0.01), block3.cells)))
block3.sigma <- crossprod(block3.p, block3.p*block3.eigens)
block3.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block3.cells, mean=5, sd=0.01), sigma=block3.sigma))
# need to through in random 0's for realism
block3.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block3.gex), byrow=TRUE)
block3.gex[block3.drop] <- 0
block4.cells <- 250
# select a set of eigen values for the covariance matrix of each block, say 50 eigenvalues?
block4.eigens <- sapply(1:n.dim, FUN=function(X) rexp(n=1, rate=abs(runif(n=1, min=0, max=50))))
block4.eigens <- block4.eigens[order(block4.eigens)]
block4.p <- qr.Q(qr(matrix(rnorm(block4.cells^2, mean=4.75, sd=0.01), block4.cells)))
block4.sigma <- crossprod(block4.p, block4.p*block4.eigens)
block4.gex <- abs(rmvnorm(n=r.n, mean=rnorm(n=block4.cells, mean=5.28, sd=0.01), sigma=block4.sigma))
# need to through in random 0's for realism
block4.drop <- matrix(sapply(1:1000, FUN=function(X) rbinom(n=500, size=1, prob=0.05) == 1), nrow=nrow(block4.gex), byrow=TRUE)
block4.gex <- sapply(1:ncol(block4.gex), FUN=function(X){
ifelse(block4.drop[, X], 0, block4.gex[, X][!block4.drop[, X]])
})
sim1.gex <- do.call(cbind, list("b1"=block1.gex, "b2"=block2.gex, "b3"=block3.gex, "b4"=block4.gex))
colnames(sim1.gex) <- paste0("Cell", 1:ncol(sim1.gex))
rownames(sim1.gex) <- paste0("Gene", 1:nrow(sim1.gex))
sim1.pca <- prcomp_irlba(t(sim1.gex), n=50, scale.=TRUE, center=TRUE)
set.seed(42)
block1.cond <- rep("A", block1.cells)
block1.a <- sample(1:block1.cells, size=floor(block1.cells*0.9))
block1.b <- setdiff(1:block1.cells, block1.a)
block1.cond[block1.b] <- "B"
block2.cond <- rep("A", block2.cells)
block2.a <- sample(1:block2.cells, size=floor(block2.cells*0.5))
block2.b <- setdiff(1:block2.cells, block2.a)
block2.cond[block2.b] <- "B"
block3.cond <- rep("A", block3.cells)
block3.a <- sample(1:block3.cells, size=floor(block3.cells*0.5))
block3.b <- setdiff(1:block3.cells, block3.a)
block3.cond[block3.b] <- "B"
block4.cond <- rep("A", block4.cells)
block4.a <- sample(1:block4.cells, size=floor(block4.cells*0.05))
block4.b <- setdiff(1:block4.cells, block4.a)
block4.cond[block4.b] <- "B"
meta.df <- data.frame("Block"=c(rep("B1", block1.cells), rep("B2", block2.cells), rep("B3", block3.cells), rep("B4", block4.cells)),
"Condition"=c(block1.cond, block2.cond, block3.cond, block4.cond),
"Replicate"=c(rep("R1", floor(block1.cells*0.33)), rep("R2", floor(block1.cells*0.33)),
rep("R3", block1.cells-(2*floor(block1.cells*0.33))),
rep("R1", floor(block2.cells*0.33)), rep("R2", floor(block2.cells*0.33)),
rep("R3", block2.cells-(2*floor(block2.cells*0.33))),
rep("R1", floor(block3.cells*0.33)), rep("R2", floor(block3.cells*0.33)),
rep("R3", block3.cells-(2*floor(block3.cells*0.33))),
rep("R1", floor(block4.cells*0.33)), rep("R2", floor(block4.cells*0.33)),
rep("R3", block4.cells-(2*floor(block4.cells*0.33)))))
colnames(meta.df) <- c("Block", "Condition", "Replicate")
# define a "sample" as teh combination of condition and replicate
meta.df$Sample <- paste(meta.df$Condition, meta.df$Replicate, sep="_")
meta.df$Vertex <- c(1:nrow(meta.df))
rownames(meta.df) <- colnames(sim1.gex)
sim1.sce <- SingleCellExperiment(assays=list(logcounts=sim1.gex),
reducedDims=list("PCA"=sim1.pca$x))
sim1.mylo <- Milo(sim1.sce)
# build a graph - this can take a while for large graphs - will need to play
# around with the parallelisation options
sim1.mylo <- buildGraph(sim1.mylo, k=30, d=10)
# define neighbourhoods - this is slow for large data sets
# how can this be sped up? There are probably some parallelisable steps
sim1.mylo <- makeNhoods(sim1.mylo, k=30, prop=0.2, refined=TRUE,
d=10,
reduced_dims="PCA")
sim1.mylo <- calcNhoodDistance(sim1.mylo, d=10)
sim1.mylo <- buildNhoodGraph(sim1.mylo)
sim1.meta <- data.frame("Condition"=c(rep("A", 3), rep("B", 3)),
"Replicate"=rep(c("R1", "R2", "R3"), 2))
sim1.meta$Sample <- paste(sim1.meta$Condition, sim1.meta$Replicate, sep="_")
rownames(sim1.meta) <- sim1.meta$Sample
sim1.mylo <- countCells(sim1.mylo, samples="Sample", meta.data=meta.df)
sim1.res <- testNhoods(sim1.mylo, design=~Condition, fdr.weighting="k-distance",
design.df=sim1.meta[colnames(nhoodCounts(sim1.mylo)), ])
sim1.res <- groupNhoods(sim1.mylo, sim1.res)
test_that("Incorrect input gives the expected errors", {
expect_error(testDiffExp(matrix(0L, nrow=nrow(sim1.mylo), ncol=ncol(sim1.mylo))),
"Unrecognised input type")
expect_error(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition, assay="junk"),
"Unrecognised assay slot")
expect_error(suppressWarnings(testDiffExp(sim1.mylo, na.res, meta.data=meta.df,
design=~Condition, na.function="junk")),
"NA function junk not recognised")
# pass a model matrix with wrong dimensions
fake.model <- model.matrix(~Condition, data=meta.df)
fake.model <- fake.model[-c(1:10), ]
expect_error(suppressWarnings(testDiffExp(sim1.mylo, sim1.res,
meta.data=meta.df,
design=fake.model)),
"Cannot subset model matrix, subsetting vector is wrong length")
})
test_that("Less than optimal input gives the expected warnings", {
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition, na.function=NULL,
)),
"NULL passed to na.function, using na.pass")
na.res <- sim1.res
na.res$SpatialFDR[1:10] <- NA
expect_warning(suppressMessages(testDiffExp(sim1.mylo, na.res, meta.data=meta.df,
design=~Condition)),
"NA values found in SpatialFDR vector")
fake.meta <- meta.df
rownames(fake.meta) <- paste0("X", c(1:nrow(meta.df)))
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=fake.meta,
design=~Condition)),
"Column names of x are not the same as meta-data rownames")
fake.model <- model.matrix(~Condition, data=meta.df)
rownames(fake.model) <- paste0("X", c(1:nrow(fake.model)))
expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=fake.model)),
"Design matrix and meta-data dimnames are not the same")
# add simulated CPMs to the data
# skip this for now until I can properly debug the issue with a working
# copy of R4.3
# ux.1 <- matrix(rpois((nrow(meta.df)*r.n)/2, 5), ncol=nrow(meta.df))
# ux.2 <- matrix(rpois((nrow(meta.df)*r.n)/2, 4), ncol=nrow(meta.df))
# ux <- rbind(ux.1, ux.2)
# ux.cpm <- apply(ux, 2, function(X) log(X/(sum(X+1)/(1e6+1))))
# colnames(ux.cpm) <- colnames(sim1.mylo)
# rownames(ux.cpm) <- rownames(sim1.mylo)
#
# SingleCellExperiment::cpm(sim1.mylo) <- ux.cpm
# expect_warning(suppressMessages(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
# design=~Condition, assay="cpm")),
# "Assay type is not counts or logcounts")
})
test_that("Output is correct type", {
full.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~Condition))
expect_identical(class(full.out) , "list")
})
test_that("Contrasts can be passed without error" , {
set.seed(42)
block1.cond <- rep("A", block1.cells)
block1.a <- sample(1:block1.cells, size=floor(block1.cells*0.05))
block1.b <- setdiff(1:block1.cells, block1.a)
block1.c <- sample(1:block1.cells, size=floor(block1.cells*0.9))
block1.cond[block1.b] <- "B"
block1.cond[block1.c] <- "C"
block2.cond <- rep("A", block2.cells)
block2.a <- sample(1:block2.cells, size=floor(block2.cells*0.3))
block2.b <- setdiff(1:block2.cells, block2.a)
block2.c <- sample(1:block2.cells, size=floor(block2.cells*0.3))
block2.cond[block2.b] <- "B"
block2.cond[block2.c] <- "C"
block3.cond <- rep("A", block3.cells)
block3.a <- sample(1:block3.cells, size=floor(block3.cells*0.3))
block3.b <- setdiff(1:block3.cells, block3.a)
block3.c <- setdiff(1:block3.cells, sample(1:block3.cells, size=floor(block3.cells*0.3)))
block3.cond[block3.b] <- "B"
block3.cond[block3.c] <- "C"
meta.df <- data.frame("Block"=c(rep("B1", block1.cells), rep("B2", block2.cells), rep("B3", block3.cells)),
"Condition"=c(block1.cond, block2.cond, block3.cond),
"Replicate"=c(rep("R1", floor(block1.cells*0.33)), rep("R2", floor(block1.cells*0.33)),
rep("R3", block1.cells-(2*floor(block1.cells*0.33))),
rep("R1", floor(block2.cells*0.33)), rep("R2", floor(block2.cells*0.33)),
rep("R3", block2.cells-(2*floor(block2.cells*0.33))),
rep("R1", floor(block3.cells*0.33)), rep("R2", floor(block3.cells*0.33)),
rep("R3", block3.cells-(2*floor(block3.cells*0.33)))))
colnames(meta.df) <- c("Block", "Condition", "Replicate")
# define a "sample" as teh combination of condition and replicate
meta.df$Sample <- paste(meta.df$Condition, meta.df$Replicate, sep="_")
meta.df$Vertex <- c(1:nrow(meta.df))
#meta.df$Condition <- ordered(meta.df$Condition, levels=c("A", "B", "C"))
rownames(meta.df) <- colnames(sim1.gex[, c(1:nrow(meta.df))])
blockC.sce <- SingleCellExperiment(assays=list(logcounts=sim1.gex[, c(1:nrow(meta.df))]),
reducedDims=list("PCA"=sim1.pca$x[c(1:nrow(meta.df)), ]))
blockC.mylo <- Milo(blockC.sce)
# build a graph - this can take a while for large graphs - will need to play
# around with the parallelisation options
blockC.mylo <- buildGraph(blockC.mylo, k=30, d=10)
# define neighbourhoods - this is slow for large data sets
# how can this be sped up? There are probably some parallelisable steps
blockC.mylo <- makeNhoods(blockC.mylo, k=30, prop=0.1, refined=TRUE,
d=10, reduced_dims="PCA")
blockC.mylo <- calcNhoodDistance(blockC.mylo, d=10)
blockC.mylo <- buildNhoodGraph(blockC.mylo)
blockC.meta <- data.frame("Condition"=c(rep("A", 3), rep("B", 3), rep("C", 3)),
"Replicate"=rep(c("R1", "R2", "R3"), 3))
blockC.meta$Sample <- paste(blockC.meta$Condition, blockC.meta$Replicate, sep="_")
rownames(blockC.meta) <- blockC.meta$Sample
blockC.mylo <- countCells(blockC.mylo, samples="Sample", meta.data=meta.df)
blockC.res <- testNhoods(blockC.mylo, design=~0 + Condition, fdr.weighting="k-distance",
design.df=blockC.meta[colnames(nhoodCounts(blockC.mylo)), ])
blockC.res <- groupNhoods(blockC.mylo, blockC.res)
# works with estimable contrast levels
expect_error(do.call(rbind.data.frame,
suppressWarnings(testDiffExp(blockC.mylo, blockC.res, meta.data=meta.df,
model.contrasts=c("ConditionC - ConditionB"),
design=~0 + Condition,
gene.offset=FALSE))),
NA)
})
test_that("Row subsetting returns expected number of results", {
full.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
design=~0 + Condition
))
full.lengths <- lapply(full.out, nrow)
subset.out <- suppressWarnings(testDiffExp(sim1.mylo, sim1.res, meta.data=meta.df,
subset.row=sample(1:nrow(sim1.mylo), size=100),
design=~0 + Condition))
subset.lengths <- lapply(subset.out, nrow)
expect_true(all(unlist(full.lengths) > unlist(subset.lengths)))
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.