Materials and methods
The Ethics Review Committee on Human Research of the University of Tartu approved the study and the participants, and their legal representatives gave prior consent to participate in the study and publish the results. Blood samples were obtained from the proband (716), her brother (715), and parents (710 and 711) from a family without previous history of OI (Fig. 1). Genomic DNA was extracted from EDTA-preserved blood according to standard high-salt extraction methods and stored at −80 °C.
Fig. 1 Pedigree structure of an Estonian family affected with type II OI. DNA was collected from father (710), mother (711), brother (715), and proband (716)
The proband (716) was the first child (first pregnancy, first delivery) of non-consanguineous Estonian parents; age of the mother and father were 23 and 27 years, respectively. The parents were healthy without history of chronic or clinically significant diseases.
The girl was born with totally soft skull, the head was large (diameter 43 cm), and bones in occipital region were not palpable (skinhead). Her head was flat, being collapsed from front to back direction. She has exophthalmia and blue sclerae. The newborn had disproportional growth retardation (limbs > body): short, bowed arms and legs, and both hips were hyperflexed and turned outward.
Total skinhead and only partially developed upper part of parietal and mandibular bones were detected in Babygram X-ray investigation on date of birth. All long bones were extremely osteopenic; she had accordion-type ribs, and several fractures in different healing stages were detected in every bone. In addition, fresh right humeral fracture and left tibial fracture were confirmed.
In genetic counseling, at the age of 3 days, the baby was diagnosed with osteogenesis imperfecta type II based on clinical signs and X-ray data. As family history was negative for the OI, de novo autosomal dominant mutation was suspected.

Exome sequencing
Whole-exome sequencing was performed on an affected child and both unaffected parents at the NGS core facility of the Estonian Genome Center, University of Tartu. Exome capture was performed using the TruSeq Exome Enrichment kit (Illumina) following the manufacturer’s protocol. The captured libraries were sequenced with Illumina HiSeq2000 with 100-bp paired-end reads. Over 10 Gb of sequence was generated from each individual, resulting in a coverage depth of 84× for both parents and 87× for an affected child. Sequence reads were aligned to the human reference genome (hg19, GRCh37) with the Burrows-Wheeler Aligner (BWA, version 0.6.1) [15]. Single-nucleotide substitutions and small indel variants were called with SAM tools (version 0.1.18), Picard tools (version 1.60), and a Genome Analysis Toolkit (GATK, version 1.5.21) [16, 17]. Genotypes were called at all positions with high-quality sequence bases and filtered to retain SNPs and insertion-deletions with Phred-like quality scores of at least 20. We focused on non-synonymous and canonical splice-site variants being absent from public datasets (including dbSNP135 and the 1000 Genomes Project) and in-house exome and full-genome data. We used the PolyPhen-2, SIFT, and Condel software tools to predict the functional effects of mutations [18–20]. Mutation analysis was performed with Sanger sequencing on an affected child (716), on both parents (710, 711), and an unaffected brother (715).