Approximately 4% of the world’s population suffers from rare diseases. A vast majority of these disorders have a genetic background. The number of genes that have been linked to human diseases is constantly growing, but there are still genetic syndromes that remain to be discovered. The diagnostic yield of genetic testing is continuously developing, and the need for testing is becoming more significant. Due to limited resources, including trained clinical geneticists, patients referred to clinical genetics units must be accurately selected. Around 30–40% of genetic disorders are associated with specific facial characteristics called dysmorphic features. As part of our research, we analyzed the performance of classifiers based on deep learning face recognition models in detecting dysmorphic features. We tested two classification problems: a multiclass problem (15 genetic disorders vs. controls) and a two-class problem (disease vs. controls). In the multiclass task, the best result reached an accuracy level of 84%. The best accuracy result in the two-class problem reached 96%. More importantly, the binary classifier detected disease features in patients with diseases that were not previously present in the training dataset. The classifier was able to generalize differences between patients and controls, and to detect abnormalities without information about the specific disorder. This indicates that a screening tool based on deep learning and facial recognition could not only detect known diseases, but also detect patients with diseases that were not previously known. In the future, this tool could help in screening patients before they are referred to the genetic unit.

Background: Despite advances in routine prenatal cytogenetic testing, most anomalous fetuses remain without a genetic diagnosis. Exome sequencing (ES) is a molecular technique that identifies sequence variants across protein-coding regions and is now increasingly used in clinical practice. Fetal phenotypes differ from postnatal and, therefore, prenatal ES interpretation requires a large amount of data deriving from prenatal testing. The aim of our study was to present initial results of the implementation of ES to prenatal diagnosis in Polish patients and to discuss its possible clinical impact on genetic counseling. Methods: In this study we performed a retrospective review of all fetal samples referred to our laboratory for ES from cooperating centers between January 2017 and June 2021. Results: During the study period 122 fetuses were subjected to ES at our institution. There were 52 abnormal ES results: 31 in the group of fetuses with a single organ system anomaly and 21 in the group of fetuses with multisystem anomalies. The difference between groups was not statistically significant. There were 57 different pathogenic or likely pathogenic variants reported in 33 different genes. The most common were missense variants. In 17 cases the molecular diagnosis had an actual clinical impact on subsequent pregnancies or other family members. Conclusions: Exome sequencing increases the detection rate in fetuses with structural anomalies and improves genetic counseling for both the affected couple and their relatives.

The aim of this study was to determine the suitability of the comparative genomic hybridization to microarray (aCGH) technique for prenatal diagnosis, but also to assess the frequency of chromosomal aberrations that may lead to fetal malformations but are not included in the diagnostic report. We present the results of the aCGH in a cohort of 7400 prenatal cases, indicated for invasive testing due to ultrasound abnormalities, high-risk for serum screening, thickened nuchal translucency, family history of genetic abnormalities or congenital abnormalities, and advanced maternal age (AMA). The overall chromosomal aberration detection rate was 27.2% (2010/7400), including 71.2% (1431/2010) of numerical aberrations and 28.8% (579/2010) of structural aberrations. Additionally, the detection rate of clinically significant copy number variants (CNVs) was 6.8% (505/7400) and 0.7% (57/7400) for variants of unknown clinical significance. The detection rate of clinically significant submicroscopic CNVs was 7.9% (334/4204) for fetuses with structural anomalies, 5.4% (18/336) in AMA, 3.1% (22/713) in the group of abnormal serum screening and 6.1% (131/2147) in other indications. Using the aCGH method, it was possible to assess the frequency of pathogenic chromosomal aberrations, of likely pathogenic and of uncertain clinical significance, in the groups of cases with different indications for an invasive test.

Congenital heart defects (CHDs) appear in 8–10 out of 1000 live born newborns and are one of the most common causes of deaths. In fetuses, the congenital heart defects are found even 3–5 times more often. Currently, microarray comparative genomic hybridization (array CGH) is recommended by worldwide scientific organizations as a first-line test in the prenatal diagnosis of fetuses with sonographic abnormalities, especially cardiac defects. We present the results of the application of array CGH in 484 cases with prenatally diagnosed congenital heart diseases by fetal ultrasound scanning (256 isolated CHD and 228 CHD coexisting with other malformations). We identified pathogenic aberrations and likely pathogenic genetic loci for CHD in 165 fetuses and 9 copy number variants (CNVs) of unknown clinical significance. Prenatal array-CGH is a useful method allowing the identification of all unbalanced aberrations (number and structure) with a much higher resolution than the currently applied traditional assessment techniques karyotype. Due to this ability, we identified the etiology of heart defects in 37% of cases.