Fig. 1

Software pipeline. The same colour bars indicate the alignment from the same reads. Half arrows indicate the orientation of the alignment. (1) The original alignment in a region. (2) Reads with multiple sub-alignments to the same chromosome are retained. Uniquely aligned reads are removed. (3) We obtain inversion signals and identify inverted repeats from sub-alignment. If inverted repeats (green dash lines) are observed, inversions are classified as NAHR, otherwise it is classified as NHEJ. Non-inversion information reads are removed. (4) Inversion signals were merged into regions as blue arcs. (5) Once the inversion regions are defined, we estimated the number of inversion reads as well as the number of reads supporting the non-inverted (reference) allele, which are removed in the step (2). Horizontal black dash line indicates the classification of inversion and reference reads. Finally, the software predicts the inversion with position, mechanism and genotype