0.1 Make plots

#tight <- read.csv("collman15v2_annotations_tight_F0.csv")
tight <- read.csv("tightRestricted_anno_collman.csv")
cubes <- read.csv("collman15v2_annotations_11cubes_F0.csv")
stight <- scale(tight, center = TRUE, scale = TRUE)
scubes <- scale(cubes, center = TRUE, scale = TRUE)

ccol <- c('blue', 'blue', 'blue', 'red', 'red', 'red', 'black', 'black', 'black', 'green', 'green', 'green')

set.seed(317)
Lt <- runAll(stight, ccol = ccol)
Lt[[1]] <- mlocation(tight, ccol = ccol)

Lc <- runAll(scubes, ccol = ccol)
Lc[[1]] <- mlocation(cubes, ccol = ccol)

w = 720
h = 720 

png("d1heat.png", width = w, height = h)
p1 <- plot(Lt[[2]]) 
p2 <- plot(Lc[[2]])
grid.arrange(p1,p2)
dev.off()

png("mlocation.png", width = 2*w, height = h)
p1 <- plot(Lt[[1]]) 
p2 <- plot(Lc[[1]])
grid.arrange(p1,p2)
dev.off()

png("cumulativeVariance.png", width = w, height = h)
p1 <- plot(Lt[[3]]) 
p2 <- plot(Lc[[3]])
grid.arrange(p1,p2)
dev.off()

png("outliers.png", width = w, height = h)
p1 <- plot(Lt[[4]]) 
p2 <- plot(Lc[[4]])
grid.arrange(p1,p2)
dev.off()

png("cor.png", width = w, height = 2*h)
par(mfrow=c(2,1))
plot(Lt[[6]]) 
plot(Lc[[6]])
dev.off()

png("pairhexTight.png", width = 2*w, height = 2*h)
pairhex(stight)
dev.off()

png("pairhexCubes.png", width = 2*w, height = 2*h)
pairhex(scubes)
dev.off()

png("hmcClassificationsTight.png", width = 2*w, height = 2*h)
cr <- viridis(max(Lt[[7]]$dat$labels$col))
pairs(Lt[[7]]$dat$data, pch = 19, cex = 0.5, col = cr[Lt[[7]]$dat$labels$col])
dev.off()

png("hmcClassificationsCubes.png", width = 2*w, height = 2*h)
cr <- viridis(max(Lc[[7]]$dat$labels$col))
pairs(Lc[[7]]$dat$data, col = cr, pch = 19, cex = 0.5)
dev.off()

png("dendrograms.png", width = w, height = h)
par(mfrow = c(2,1))
plotDend(Lt[[7]])
plotDend(Lc[[7]])
dev.off()

png("stackMeans.png", width = w, height = 2*h)
p1 <- stackM(Lt[[7]], ccol = ccol, depth = 3, centered = TRUE)
p2 <- stackM(Lc[[7]], ccol = ccol, depth = 2, centered = TRUE)
grid.arrange(p1,p2)
dev.off()

png("clusterMeans.png", width = w, height = 0.75*h)
p1 <- clusterMeans(Lt[[7]], ccol = ccol)
p2 <- clusterMeans(Lc[[7]], ccol = ccol)
grid.arrange(p1,p2)
dev.off()

1 Tight F0 plots are on top, 11x11x11 F0 plots are on the bottom

1.1 1-d Heatmap

1.2 Location meda_plots

1.3 Outliers as given by randomForest

1.4 Correlation Matrix

1.5 Cumulative Variance with Elbows

1.6 Paired Hex-binned plot

1.7 Hierarchical GMM Classifications

1.8 Hierarchical GMM Dendrogram

1.9 Stacked Means

1.10 Cluster Means