tests/testthat/test_buildSpatialGraph.R
In BodenmillerGroup/imcRtools: Methods for imaging mass cytometry data analysis
test_that("buildSpatialGraph function works", {
library(cytomapper)
data("pancreasSCE")
cur_sce1 <- pancreasSCE[,pancreasSCE$ImageNb == 1]
cur_sce2 <- pancreasSCE[,pancreasSCE$ImageNb == 2]
cur_sce3 <- pancreasSCE[,pancreasSCE$ImageNb == 3]
cur_sce1$Pos_X <- cur_sce1$Pos_X - min(cur_sce1$Pos_X)
cur_sce1$Pos_Y <- cur_sce1$Pos_Y - min(cur_sce1$Pos_Y)
cur_sce2$Pos_X <- cur_sce2$Pos_X - min(cur_sce2$Pos_X)
cur_sce2$Pos_Y <- cur_sce2$Pos_Y - min(cur_sce2$Pos_Y)
cur_sce3$Pos_X <- cur_sce3$Pos_X - min(cur_sce3$Pos_X)
cur_sce3$Pos_Y <- cur_sce3$Pos_Y - min(cur_sce3$Pos_Y)
pancreasSCE <- cbind(cur_sce1, cur_sce2, cur_sce3)
# Delauney
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay"),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2082)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 11, 13, 50, 4, 6, 10, 12, 1, 2, 3))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay", directed = FALSE)
expect_equal(length(colPair(cur_sce)), 2082)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 11, 13, 50, 4, 6, 10, 12, 1, 2, 3))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
# Max dist
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay", max_dist = 20),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 1956)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(3L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L, 5L))
expect_equal(to(colPair(cur_sce))[10:20], c(12L, 3L, 7L, 8L, 10L, 2L, 9L, 13L, 14L, 19L, 24L))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
# KNN
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 1810)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(22, 4, 6, 10, 12, 17, 3, 7, 8, 10, 16))
cur_ks <- table(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_ks <- rowSums(cur_ks)
expect_true(all(cur_ks == 5))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2068)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(22, 4, 6, 10, 12, 17, 3, 7, 8, 10, 16))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Other ks
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 3620)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
cur_ks <- table(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_ks <- rowSums(cur_ks)
expect_true(all(cur_ks == 10))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 4124)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Other algorithm
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "vptree",
BNPARAM = BiocNeighbors::VptreeParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "annoy",
BNPARAM = BiocNeighbors::AnnoyParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "hnsw",
BNPARAM = BiocNeighbors::HnswParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce),
c("knn_interaction_graph", "vptree", "annoy", "hnsw"))
expect_equal(colPair(cur_sce, "knn_interaction_graph"),
colPair(cur_sce, "vptree"))
# Max dist
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5, max_dist = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 711)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(9, 9, 10, 10, 11, 12, 12, 12, 13, 13, 13))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 14, 4, 7, 13, 3, 17, 18, 5, 11, 19))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Expansion
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2042)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 6))
expect_equal(to(colPair(cur_sce))[10:20], c(3, 7, 8, 10, 2, 11, 13, 19, 3, 9, 12))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2042)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 6))
expect_equal(to(colPair(cur_sce))[10:20], c(3, 7, 8, 10, 2, 11, 13, 19, 3, 9, 12))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 1),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 0)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
# SpatialExperiment
library(SpatialExperiment)
pancreasSPE <- SpatialExperiment(assays = list(counts = counts(pancreasSCE)),
sample_id = pancreasSCE$ImageName)
colData(pancreasSPE) <- colData(pancreasSCE)
colData(pancreasSPE)[,c("Pos_X", "Pos_Y")] <- NULL
colPairs(pancreasSPE) <- colPairs(pancreasSCE)
expect_error(cur_sce <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "'coords' not in spatialCoords(object).",
fixed = TRUE)
spatialCoords(pancreasSPE) <- as.matrix(colData(pancreasSCE)[,c("Pos_X", "Pos_Y")])
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "expansion_interaction_graph")
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "delaunay"),
regex = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "delaunay_interaction_graph")
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "knn", k = 5),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "knn_interaction_graph")
# Check that metadata is not duplicated
cur_sce <- pancreasSCE
metadata(cur_sce) <- list(test = c(1,2))
cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 5)
expect_equal(length(metadata(cur_sce)), 1)
expect_equal(metadata(cur_sce), list(test = c(1, 2)))
int_metadata(cur_sce) <- list(version = "1.16.0")
cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 5)
expect_equal(length(int_metadata(cur_sce)), 1)
expect_equal(int_metadata(cur_sce), list(version = "1.16.0"))
# Fail
expect_error(buildSpatialGraph("test"),
regexp = "'object' not of type 'SingleCellExperiment'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE),
regexp = "argument \"img_id\" is missing, with no default",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = 1),
regexp = "'img_id' must be a single string.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "test"),
regexp = "'img_id' not in colData(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion"),
regexp = "When constructing a graph via expansion, please specify 'threshold'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion"),
regexp = "When constructing a graph via expansion, please specify 'threshold'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion",
threshold = "test"),
regexp = "'threshold' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn"),
regexp = "When constructing a graph via nearest neighbour detection, please specify 'k'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn",
k = "test"),
regexp = "'k' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn",
k = 10, max_dist = "test"),
regexp = "'max_dist' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = "test"),
regexp = "'coords' must be a character vector of length 2.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = c(2, 1)),
regexp = "'coords' must be a character vector of length 2.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = c("Pos_X", "test")),
regexp = "'coords' not in colData(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSPE, img_id = "ImageNb", type = "delaunay",
coords = c("Pos_X", "test")),
regexp = "'coords' not in spatialCoords(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
name = 1),
regexp = "'name' must be a single string.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
directed = 1),
regexp = "'directed' must be a single logical.",
fixed = TRUE)
})
test_that("buildSpatialGraph function works", {
skip_on_os(os = "windows")
library(cytomapper)
data("pancreasSCE")
# Parallelisation
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15,
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay",
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5,
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
})
BodenmillerGroup/imcRtools documentation built on Oct. 30, 2024, 12:19 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.
test_that("buildSpatialGraph function works", {
library(cytomapper)
data("pancreasSCE")
cur_sce1 <- pancreasSCE[,pancreasSCE$ImageNb == 1]
cur_sce2 <- pancreasSCE[,pancreasSCE$ImageNb == 2]
cur_sce3 <- pancreasSCE[,pancreasSCE$ImageNb == 3]
cur_sce1$Pos_X <- cur_sce1$Pos_X - min(cur_sce1$Pos_X)
cur_sce1$Pos_Y <- cur_sce1$Pos_Y - min(cur_sce1$Pos_Y)
cur_sce2$Pos_X <- cur_sce2$Pos_X - min(cur_sce2$Pos_X)
cur_sce2$Pos_Y <- cur_sce2$Pos_Y - min(cur_sce2$Pos_Y)
cur_sce3$Pos_X <- cur_sce3$Pos_X - min(cur_sce3$Pos_X)
cur_sce3$Pos_Y <- cur_sce3$Pos_Y - min(cur_sce3$Pos_Y)
pancreasSCE <- cbind(cur_sce1, cur_sce2, cur_sce3)
# Delauney
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay"),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2082)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 11, 13, 50, 4, 6, 10, 12, 1, 2, 3))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay", directed = FALSE)
expect_equal(length(colPair(cur_sce)), 2082)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 11, 13, 50, 4, 6, 10, 12, 1, 2, 3))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
# Max dist
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay", max_dist = 20),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 1956)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "delaunay_interaction_graph")
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(3L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L, 5L))
expect_equal(to(colPair(cur_sce))[10:20], c(12L, 3L, 7L, 8L, 10L, 2L, 9L, 13L, 14L, 19L, 24L))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
# KNN
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 1810)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(22, 4, 6, 10, 12, 17, 3, 7, 8, 10, 16))
cur_ks <- table(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_ks <- rowSums(cur_ks)
expect_true(all(cur_ks == 5))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2068)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4))
expect_equal(to(colPair(cur_sce))[10:20], c(22, 4, 6, 10, 12, 17, 3, 7, 8, 10, 16))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Other ks
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 3620)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
cur_ks <- table(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_ks <- rowSums(cur_ks)
expect_true(all(cur_ks == 10))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){which(x == 1)})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 4124)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:11])})
cur_ind_real <- graph_from_adj_list(as.list(as.data.frame(cur_ind_real)))
cur_ind_real <- as.undirected(cur_ind_real)
cur_ind_real <- as.directed(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Other algorithm
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "vptree",
BNPARAM = BiocNeighbors::VptreeParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "annoy",
BNPARAM = BiocNeighbors::AnnoyParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_message(cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 10, name = "hnsw",
BNPARAM = BiocNeighbors::HnswParam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce),
c("knn_interaction_graph", "vptree", "annoy", "hnsw"))
expect_equal(colPair(cur_sce, "knn_interaction_graph"),
colPair(cur_sce, "vptree"))
# Max dist
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5, max_dist = 10),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 711)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "knn_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(9, 9, 10, 10, 11, 12, 12, 12, 13, 13, 13))
expect_equal(to(colPair(cur_sce))[10:20], c(6, 14, 4, 7, 13, 3, 17, 18, 5, 11, 19))
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){sort(order(x)[2:6])})
cur_dist_real <- apply(as.matrix(cur_dist), 1, function(x){x[sort(order(x)[2:6])]})
cur_ind_real <- lapply(1:ncol(cur_ind_real), function(x){
cur_ind_real[cur_dist_real[,x] <= 10,x]
})
cur_ind_real <- cur_ind_real[lapply(cur_ind_real, length) > 0]
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
cur_ind_test <- apply(cur_ind_test, 1, function(x){as.numeric(names(which(x == 1)))})
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
# Expansion
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2042)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 6))
expect_equal(to(colPair(cur_sce))[10:20], c(3, 7, 8, 10, 2, 11, 13, 19, 3, 9, 12))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15, directed = FALSE),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 2042)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
expect_equal(from(colPair(cur_sce))[10:20], c(4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 6))
expect_equal(to(colPair(cur_sce))[10:20], c(3, 7, 8, 10, 2, 11, 13, 19, 3, 9, 12))
cur_edges_1 <- paste(from(colPair(cur_sce)), to(colPair(cur_sce)))
cur_edges_2 <- paste(to(colPair(cur_sce)), from(colPair(cur_sce)))
expect_equal(sort(cur_edges_1), sort(cur_edges_2))
expect_equal(sum(isRedundantHit(colPair(cur_sce))), length(colPair(cur_sce))/2)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 1]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 2]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
cur_sce_1 <- cur_sce[,cur_sce$ImageNb == 3]
cur_dist <- dist(as.matrix(colData(cur_sce_1)[,c("Pos_X", "Pos_Y")]))
cur_ind_real <- apply(as.matrix(cur_dist), 1, function(x){which(x <= 15)})
cur_ind_real <- graph_from_adj_list(cur_ind_real)
cur_ind_real <- simplify(cur_ind_real)
cur_ind_real <- as_edgelist(cur_ind_real)
cur_ind_real <- table(cur_ind_real[,1], cur_ind_real[,2])
cur_ind_test <- table(from(colPair(cur_sce_1)), to(colPair(cur_sce_1)))
expect_equal(cur_ind_real, cur_ind_test, check.attributes = FALSE)
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 1),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_true(!is.null(colPair(cur_sce)))
expect_equal(length(colPair(cur_sce)), 0)
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph")
expect_silent(print(p))
p <- plotSpatial(cur_sce, img_id = "ImageNb", draw_edges = TRUE,
colPairName = "expansion_interaction_graph",
directed = FALSE)
expect_silent(print(p))
# SpatialExperiment
library(SpatialExperiment)
pancreasSPE <- SpatialExperiment(assays = list(counts = counts(pancreasSCE)),
sample_id = pancreasSCE$ImageName)
colData(pancreasSPE) <- colData(pancreasSCE)
colData(pancreasSPE)[,c("Pos_X", "Pos_Y")] <- NULL
colPairs(pancreasSPE) <- colPairs(pancreasSCE)
expect_error(cur_sce <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "'coords' not in spatialCoords(object).",
fixed = TRUE)
spatialCoords(pancreasSPE) <- as.matrix(colData(pancreasSCE)[,c("Pos_X", "Pos_Y")])
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "expansion", threshold = 15),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "expansion_interaction_graph")
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "delaunay"),
regex = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "delaunay_interaction_graph")
expect_message(cur_spe <- buildSpatialGraph(pancreasSPE, img_id = "ImageNb",
type = "knn", k = 5),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_spe), "knn_interaction_graph")
# Check that metadata is not duplicated
cur_sce <- pancreasSCE
metadata(cur_sce) <- list(test = c(1,2))
cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 5)
expect_equal(length(metadata(cur_sce)), 1)
expect_equal(metadata(cur_sce), list(test = c(1, 2)))
int_metadata(cur_sce) <- list(version = "1.16.0")
cur_sce <- buildSpatialGraph(cur_sce, img_id = "ImageNb",
type = "knn", k = 5)
expect_equal(length(int_metadata(cur_sce)), 1)
expect_equal(int_metadata(cur_sce), list(version = "1.16.0"))
# Fail
expect_error(buildSpatialGraph("test"),
regexp = "'object' not of type 'SingleCellExperiment'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE),
regexp = "argument \"img_id\" is missing, with no default",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = 1),
regexp = "'img_id' must be a single string.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "test"),
regexp = "'img_id' not in colData(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion"),
regexp = "When constructing a graph via expansion, please specify 'threshold'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion"),
regexp = "When constructing a graph via expansion, please specify 'threshold'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "expansion",
threshold = "test"),
regexp = "'threshold' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn"),
regexp = "When constructing a graph via nearest neighbour detection, please specify 'k'.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn",
k = "test"),
regexp = "'k' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "knn",
k = 10, max_dist = "test"),
regexp = "'max_dist' must be a single numeric",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = "test"),
regexp = "'coords' must be a character vector of length 2.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = c(2, 1)),
regexp = "'coords' must be a character vector of length 2.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
coords = c("Pos_X", "test")),
regexp = "'coords' not in colData(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSPE, img_id = "ImageNb", type = "delaunay",
coords = c("Pos_X", "test")),
regexp = "'coords' not in spatialCoords(object).",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
name = 1),
regexp = "'name' must be a single string.",
fixed = TRUE)
expect_error(buildSpatialGraph(pancreasSCE, img_id = "ImageNb", type = "delaunay",
directed = 1),
regexp = "'directed' must be a single logical.",
fixed = TRUE)
})
test_that("buildSpatialGraph function works", {
skip_on_os(os = "windows")
library(cytomapper)
data("pancreasSCE")
# Parallelisation
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "expansion", threshold = 15,
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "expansion_interaction_graph")
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "delaunay",
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "delaunay_interaction_graph")
expect_message(cur_sce <- buildSpatialGraph(pancreasSCE, img_id = "ImageNb",
type = "knn", k = 5,
BPPARAM = BiocParallel::bpparam()),
regexp = "The returned object is ordered by the 'ImageNb' entry.")
expect_equal(colPairNames(cur_sce), "knn_interaction_graph")
})
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.