Genome sequencing in families with congenital limb malformations

Abstract

The extensive clinical and genetic heterogeneity of congenital limb malformation calls for comprehensive genome-wide analysis of genetic variation. Genome sequencing (GS) has the potential to identify all genetic variants. Here we aim to determine the diagnostic potential of GS as a comprehensive one-test-for-all strategy in a cohort of undiagnosed patients with congenital limb malformations. We collected 69 cases (64 trios, 1 duo, 5 singletons) with congenital limb malformations with no molecular diagnosis after standard clinical genetic testing and performed genome sequencing. We also developed a framework to identify potential noncoding pathogenic variants. We identified likely pathogenic/disease-associated variants in 12 cases (17.4%) including four in known disease genes, and one repeat expansion in HOXD13. In three unrelated cases with ectrodactyly, we identified likely pathogenic variants in UBA2, establishing it as a novel disease gene. In addition, we found two complex structural variants (3%). We also identified likely causative variants in three novel high confidence candidate genes. We were not able to identify any noncoding variants. GS is a powerful strategy to identify all types of genomic variants associated with congenital limb malformation, including repeat expansions and complex structural variants missed by standard diagnostic approaches. In this cohort, no causative noncoding SNVs could be identified.

Fig. 1

Inversion-deletion at SHFM3 locus: a pedigree, N.T. not tested. b feet of grand-uncle (II-3). c hands and feet of the index patient (IV-1). d genomic architecture of SHFM3. e GS data of the family, note the presence of an inversion (chr10: 103,321,526–103,426,609) flanked by deletions (chr10:103,319,219–103,321,525 and chr10:103,426,610–103,436,718) on either site

Fig. 2

a Pedigree and phenotype of individual I11. b Potential neo-TAD at the fusion site. c Breakpoint and fusion sites between regions from chr7 and chr9

Fig. 3

UBA2 variants and ectrodactyly. a–c Patients with likely pathogenic UBA2 variants upper panels: pedigrees, N.T.: not tested; middle panels: characteristic limb malformations, lower panels: sequencing data. d conservation of Asp50 mutated in individual I14, numbers indicate amino acid residues, yellow bars highlight positions tested by Olsen et al. to cause loss of function when substituted by alanine (Olsen et al. 2010)

Fig. 4

Pipeline of noncoding data analysis