Analysis of a repetitive transgene reveals inversions and deletions that contribute to the generation of aberrant RNAs. (A) Schematic of the repeat structure of the sur-5::gfp transgene. Repeating units of the ∼12-kb plasmid, which has sur-5 (gray box) and gfp (black box) coding regions, deduced from Illumina sequencing, and units that are not schematized (…) are indicated. Also see Fig. S1. (B) DNA rearrangements in the sur-5::gfp transgene. Inversions (blue) and deletions (orange) observed by DNA sequencing (DNA-seq) are represented on one of the units of the sur-5::gfp repetitive transgene. Percentages reflect reads supporting each rearrangement at a given position compared with those that support no rearrangements at that position. (C) Microhomology-dependent generation of inversions and deletions from a circular plasmid could explain rearrangements found in sur-5::gfp. (top) Schematic illustrating how circular plasmids could undergo inversions (left) or deletions (right) at sites of sequence microhomology (xyz) during the formation of arrays. Dotted lines indicate proposed sites of rearrangements. (bottom) Rearrangements found in sur-5::gfp are associated with regions of microhomology. Identical sequences present near the sites of rearrangement are indicated for inversions (a–r as in B) and deletions (s–w as in B). Dotted lines and sequence colors (brown and blue) are as in the top panel. X indicates nonidentical bases. (D) Rearrangements in RNAs made from the sur-5::gfp transgene. Inversions and deletions (or alternative splicing) observed by RNA sequencing (RNA-seq) and percentages are as in B.