Search

Additional file 1: Figure S1. T2T-CHM13 inversion callset summary and comparison to GRCh38 (n = 373). Figure S2. Differences between GRCh38 and T2T-CHM13 callsets. Figure S3. Inversion callset summary with respect to T2T-CHM13 reference. Figure S4. Nonsyntenic and likely novel sites in T2T-CHM13 inversion calls. Figure S5. Enrichment of inversions in pericentromeric regions. Figure S6. Sequence composition of inversions from pericentromeric regions. Figure S7. Novel pericentromeric inversion on chromosome 1. Figure S8. Complete assemblies of chromosome 1 centromeric region. Figure S9. Relative position of alpha satellite array and novel pericentromeric inversion on chromosome 1. Figure S10. Inversion phasing at pericentromeric region of chromosome 7. Figure S11. Evaluation of putative misorients in GRCh38 with respect to T2T-CHM13. Figure S12. Evaluation of inversion differences between GRCh38 and T2T-CHM13 references. Figure S13. Examples of minor and misoriented alleles at chromosome 16. Figure S14. Structural differences at Xq28 between GRCh38 and T2T-CHM13. Figure S15. Diverse structural haplotypes at the Xq28 region. Figure S16. Structural differences at 16p12.2 between GRCh38 and T2T-CHM13. Figure S17. Topological differences at 16p12.2 between GRCh38 and T2T-CHM13. Figure S18. Rare inversions at disease relevant loci. Figure S19. Diverse structural haplotypes at 15q25.2 region. Figure S20. Assembled inversion breakpoints at 15q25.2 and inversion breakpoint mapping. Figure S21. Example of long-lasting misorientation errors in previous human genome references. Supplementary Notes. Consortia.

Individuals of one species share the bulk of their genetic material, yet no two genomes are the same. Aside from displaying classical variation such as deletions, insertions, or substitutions of base pairs, two DNA segments can also differ in their orientation relative to the rest of their chromosomes. Such inversions are known for a range of biological implications and contribute critically to genome evolution and disease. However, inversions are notoriously challenging to detect, a fact which still impedes comprehensive analysis of their specific properties. This thesis describes several highly inter-connected projects aimed at identifying and functionally characterizing inversions present in the human population and related great ape species. First, inversions between human and four great ape species were assessed for their potential to disrupt topologically associating domains (TADs), potentially prompting gene misregulation. TAD boundaries co-located with breakpoints of long inversions, and while disrupted TADs displayed elevated rates of differen- tially expressed genes, this effect could be attributed the vicinity to inversion breakpoints, suggesting overall robustness of gene expression in response to TAD disruption. The second part of this thesis describes contributions to a collaborative project aimed at characterizing the full spectrum of inversions in 43 humans. In this study, I co-developed a novel inversion genotyping algorithm based on Strand- specific DNA sequencing and contributed to the description of 398 inversion polymorphisms. Inversions exhibited various underlying formation mechanisms, promotion of gene dysregulation, widespread recurrence, and association with genomic disease. These results suggest that long inversions are much more prominent in humans than previously thought, with at least 0.6% of the genome subject to inversion recurrence and, sometimes, the associated risk of subsequent deleterious mutation. With a focus on the link between inversions and disease-causing copy num- ber variations, the last project describes a novel algorithm to identify loci hit sequentially by several overlapping mutation events. This algorithm enabled the description of detailed mutation sequences in 20 highly dynamic regions in the human genome, and additional complex variants on chromosome Y. Six complex loci associate directly with a genomic disease, thereby highlighting in detail the intrinsic link between inversions and CNVs. In summary, these projects provide novel insights into the landscape of in- versions in humans and primates, which are much more frequent, and often more complex than previously thought. These findings provide a basis for future inversion studies and highlight the crucial contribution of this class of mutation to genome variation.

Additional file 3. Review history.

Unlike copy number variants (CNVs), inversions remain an underexplored genetic variation class. By integrating multiple genomic technologies, we discover 729 inversions in 41 human genomes. Approximately 85% of inversions -4 per locus and generation. Recurrent inversions exhibit a sex- chromosomal bias, and significantly co-localize to the critical regions of genomic disorders. We propose that inversion recurrence results in an elevated number of heterozygous carriers and structural SD diversity, which increases mutability in the population and predisposes to disease- causing CNVs.