Development of a comprehensive noninvasive prenatal test.

Our aim was to develop and apply a comprehensive noninvasive prenatal test (NIPT) by using high-coverage targeted next-generation sequencing to estimate fetal fraction, determine fetal sex, and detect trisomy and monogenic disease without parental genotype information. We analyzed 45 pregnancies, 40 mock samples, and eight mother-child pairs to generate 35 simulated datasets. Fetal fraction (FF) was estimated based on analysis of the single nucleotide polymorphism (SNP) allele fraction distribution. A Z-score was calculated for trisomy of chromosome 21 (T21), and fetal sex detection. Monogenic disease detection was performed through variant analysis. Model validation was performed using the simulated datasets. The novel model to estimate FF was robust and accurate (r2= 0.994, p-value < 2.2e-16). For samples with FF > 0.04, T21 detection had 100% sensitivity (95% CI: 63.06 to 100%) and 98.53% specificity (95% CI: 92.08 to 99.96%). Fetal sex was determined with 100% accuracy. We later performed a proof of concept for monogenic disease diagnosis of 5/7 skeletal dysplasia cases. In conclusion, it is feasible to perform a comprehensive NIPT by using only data from high coverage targeted sequencing, which, in addition to detecting trisomies, also make it possible to identify pathogenic variants of the candidate genes for monogenic diseases.

Complexity of the 5' Untranslated Region of EIF4A3, a Critical Factor for Craniofacial and Neural Development.

Repeats in coding and non-coding regions have increasingly been associated with many human genetic disorders, such as Richieri-Costa-Pereira syndrome (RCPS). RCPS, mostly characterized by midline cleft mandible, Robin sequence and limb defects, is an autosomal-recessive acrofacial dysostosis mainly reported in Brazilian patients. This disorder is caused by decreased levels of EIF4A3, mostly due to an increased number of repeats at the EIF4A3 5'UTR. EIF4A3 5'UTR alleles are CG-rich and vary in size and organization of three types of motifs. An exclusive allelic pattern was identified among affected individuals, in which the CGCA-motif is the most prevalent, herein referred as "disease-associated CGCA-20nt motif." The origin of the pathogenic alleles containing the disease-associated motif, as well as the functional effects of the 5'UTR motifs on EIF4A3 expression, to date, are entirely unknown. Here, we characterized 43 different EIF4A3 5'UTR alleles in a cohort of 380 unaffected individuals. We identified eight heterozygous unaffected individuals harboring the disease-associated CGCA-20nt motif and our haplotype analyses indicate that there are more than one haplotype associated with RCPS. The combined analysis of number, motif organization and haplotypic diversity, as well as the observation of two apparently distinct haplotypes associated with the disease-associated CGCA-20nt motif, suggest that the RCPS alleles might have arisen from independent unequal crossing-over events between ancient alleles at least twice. Moreover, we have shown that the number and sequence of motifs in the 5'UTR region is associated with EIF4A3 repression, which is not mediated by CpG methylation. In conclusion, this study has shown that the large number of repeats in EIF4A3 does not represent a dynamic mutation and RCPS can arise in any population harboring alleles with the CGCA-20nt motif. We also provided further evidence that EIF4A3 5'UTR is a regulatory region and the size and sequence type of the repeats at 5'UTR may contribute to clinical variability in RCPS.

Targeted molecular investigation in patients within the clinical spectrum of Auriculocondylar syndrome.

Auriculocondylar syndrome, mainly characterized by micrognathia, small mandibular condyle, and question mark ears, is a rare disease segregating in an autosomal dominant pattern in the majority of the families reported in the literature. So far, pathogenic variants in PLCB4, GNAI3, and EDN1 have been associated with this syndrome. It is caused by a developmental abnormality of the first and second pharyngeal arches and it is associated with great inter- and intra-familial clinical variability, with some patients not presenting the typical phenotype of the syndrome. Moreover, only a few patients of each molecular subtype of Auriculocondylar syndrome have been reported and sequenced. Therefore, the spectrum of clinical and genetic variability is still not defined. In order to address these questions, we searched for alterations in PLCB4, GNAI3, and EDN1 in patients with typical Auriculocondylar syndrome (n = 3), Pierre Robin sequence-plus (n = 3), micrognathia with additional craniofacial malformations (n = 4), or non-specific auricular dysplasia (n = 1), which could represent subtypes of Auriculocondylar syndrome. We found novel pathogenic variants in PLCB4 only in two of three index patients with typical Auriculocondylar syndrome. We also performed a detailed comparative analysis of the patients presented in this study with those previously published, which showed that the pattern of auricular abnormality and full cheeks were associated with molecularly characterized individuals with Auriculocondylar syndrome. Finally, our data contribute to a better definition of a set of parameters for clinical classification that may be used as a guidance for geneticists ordering molecular testing for Auriculocondylar syndrome. © 2017 Wiley Periodicals, Inc.