Figure 4:

a. Dendrogram representation of hierarchical clustering applied on the SPIT DTU matrix for schizophrenia samples b. SPIT-Chart for the schizophrenia analysis: For each transcript that passed the initial filtering steps, the median $p$-value that has been observed through 1000 iterations of the SPIT-Test (median $\left({\bigcup }_S{p}_{j,s}\right)$) is plotted on the $x$-axis, and the $p$-value observed in the actual comparison of the schizophrenia samples to the controls is plotted on the $y$-axis, both on $-{log}_{10}$ scale. c. Box plots of permutation importance scores (generated from 100 permutations) of the SPIT output vector and provided covariates for the final 6 DTU genes.