1 In/Ex

I start with the Gaussian weighted averages over the 11 x 11 x 11 x Ch cubes at each synapse location.

Here I re-factor the data into a two-variable dataset:

INH \(:= \text{Z-score} \circ \sum{\text{Inhibitory Features}}\)

EX \(:= \text{Z-score} \circ \sum{\text{Excitatory Features}}\)

Then plot the new dataset colored by the given gaba labels. The ellipses superimposed on the plot correspond to the class conditional covariances (at 1,2, and 3 standard deviations) of the components.

2 Results

2.1 1-d Heatmap

2.2 Location meda_plots

2.3 Outliers as given by randomForest

2.4 Correlation Matrix

2.5 Cumulative Variance with Elbows

2.6 Paired Hex-binned plot

2.7 Hierarchical GMM Classifications

2.8 Hierarchical GMM Dendrogram

2.9 Stacked Means

2.10 Cluster Means

3 Restricting hGMM to \(K = 2\)

Here we are restricting hierarchical GMM to only go through on level. We are comparing the cluster results to the gaba labels.

set.seed(3144)
h2 <- hmc(sndat[, -1], maxDepth = 2, ccol = ccol, model = c("VVV"))
h2lab <- viridis(max(h2$dat$labels$col))
h2col <- h2$dat$labels$col

3.1 K = 2 stacked means plot

p1 <- stackM(h2, ccol = "black", centered = TRUE, depth = 1)
L1 <- h2$dat$labels$L1
p2 <- stackMraw(as.data.frame(sdat), L1, ccol = ccol, depth = 2, centered = TRUE)
grid.arrange(p1, p2, nrow = 1)

3.2 Pairs plot colored by true gaba classification

cols <- c("black", "magenta")[gabaID$gaba+1]
acols <- alpha(cols, 0.35)
#pairs(h2$dat$data, pch = 19, cex = 0.7, col = acols)
#plot(h2$dat$data, col = acols, pch = c(19,3)[gaba+1], cex = c(0.5,1)[gaba+1])
pairs(sdat, col = acols, pch = c(19,3)[gaba+1], cex = c(0.5,1)[gaba+1])

3.3 Pairs plot colored by hGMM cluster classification

acols2 <- alpha(h2lab[h2$dat$labels$col], 0.45)
par(bg = "gray45")
#plot(h2$dat$data, pch = c(3,20)[gaba + 1], cex = 1, col = acols2)
pairs(sdat, pch = 19, cex = 0.7, col = acols2)

dev.off()
## quartz 
##      2

4 Permutation test for ARI

p0 <- mclust::adjustedRandIndex(pred, gaba)
perms <- foreach(i = 1:1.5e4, .combine = c) %dopar% {
  set.seed(i*2)
  mclust::adjustedRandIndex(sample(pred), gaba)
}
pv0 <- sum(c(perms,p0) >= p0)/length(perms)
hist(perms, xlim = c(min(perms), p0 + 0.25*p0),
     main = "permutation test of ARI values", probability = TRUE)
#hist(perms, probability = TRUE)
abline(v = p0, col = 'red')

t1
##        truth
## pred    FALSE TRUE
##   FALSE   647   15
##   TRUE     61   70

5 Summary Table

measurment value
Misclassification Rate 0.0958386
Accuracy 0.9041614
Sensitivity 0.8235294
Specificity 0.9138418
Precision 0.5343511
Recall 0.8235294
ARI 0.5205843
\(p\)-value for ARI 0.000067
F1-score 0.6481481
TP 70
FP 61
TN 647
FN 15