Fig. 2

Density plots between whole genome (red) and orthologs (light blue) for Exons per Transcript (EpT) complexity metrics with densities on the y-axis and the number of exons per transcript (EpT) on x-axis. Panels are truncated to 30 exons for purposes of visualization. We observe a significant difference in EpT for whole -transcriptome annotations and partitioned by ortholog annotations in each of four example species using Wilcoxon rank sum test. A) H. sapiens (GCF_000001405.39_GRCh38.p13), W = 1.448 × 1010, P = 0, B) Drosophila melanogaster (dmel-all-r6.07), W = 5.654 × 109, P = 6.9676 × 10–106, C) Zea mays (GCF_902167145.1_Zm-B73-REFERENCE-NAM-5.0), W = 2.391 × 109, P = 0, and D) Neocallimastix californiae (GCA_002104975.1_Neocallimastix_sp._G1_v1.0_genomic), W = 1.721 × 1010, P = 7.166 × 10–118. Across all taxa, ortholog conditioning results in data that are less likely to include low complexity transcripts