Table 3.

Summary Statistics for the eQTL and DE Network

Cluster	Total	eQTL	LOAD	Top Process	Score	Count	p Value
1	8%	16%	7%	∗organ morphogenesis	1.72	8	2.900 × 10−3
2	4%	42%	34%	zinc finger, C2H2 type	0.99	8	4.100 × 10−3
3	6%	40%	12%	cell growth	1.12	5	2.400 × 10−2
4	20%	33%	16%	∗negative regulation of transcription, DNA dependent	2.05	13	2.400 × 10−3
5	6%	44%	19%	protein kinase cascade	2.06	9	2.200 × 10−3
6	9%	12%	9%	KRAB box	2.11	9	5.500 × 10−5
7	19%	21%	10%	ligase activity, forming carbon-nitrogen bonds	1.78	12	6.400 × 10−3
8	14%	35%	15%	mitochondrion	3.02	33	8.200 × 10−6
9	14%	19%	10%	glycoprotein	3.42	49	3.900 × 10−6

Shown are the proportion of transcripts, out of the total number of transcripts networked, that map to each cluster (Total), that had a significant (corrected p value < 0.05 after 1000 permutations) correlation with allele dose (eQTL), and that had significant interaction term in the 2 df model (LOAD). Top processes were defined as ontology groups that had the lowest p value within the annotation cluster with the highest enrichment score. The exception to this definition is indicated by an asterisk; for clusters 1 and 4, all of the ontology groups within the top enrichment-score bin were the same for these two clusters (see Table S2B for ontology groupings in this set). Therefore, for these clusters, top processes are listed within the next most significant grouping of ontologies. Enrichment scores (Score), numbers of transcripts within the top annotation cluster (Count), and p values for the EASE scores for each process (p Value) are given. Table S2A lists all top ontologies for each cluster.