Prioritization of putatively detrimental variants in euploid miscarriages

Abstract

STUDY QUESTION: Can small genetic variants detected in the whole genome sequencing of spontaneously aborted euploid embryos give insight into possible causes of pregnancy loss? SUMMARY ANSWER: By filtering and prioritizing genetic variants it is possible to identify genomic variants putatively responsible for miscarriage. WHAT IS KNOWN ALREADY: Miscarriage is often caused to chromosomal aneuploidies of the gametes but it can also have other genetic causes like small mutations, both de novo or inherited from parents. The analysis of genomic sequences of miscarried embryos has mostly focused on rare variation, and been carried out using criteria and methods that are difficult to reproduce. The role of small mutations has been scantily investigated so far. STUDY DESIGN, SIZE, DURATION: This is a monocentric observational study. The study includes the data analysis of 46 embryos obtained from women experiencing pregnancy loss recruited by the University of Ferrara from 2017 to 2018. The study wasapproved by the Ethical committee of Emilia-Romagna (CE/FE 170475). PARTICIPANTS/MATERIALS, SETTING, METHODS: The participants are forty-six women, mostly European (87%) diagnosed with first (n=25, av.age 32.7) or recurrent(n=21, av.age 36.5) miscarriage. Embryonic DNA was prepared form chorionic villi and used to select euploid embryos using quantitative PCR, comparative genomic hybridiztion and shallow sequencing of random genomic regions. Euploid embryos were whole-genome sequenced at 30X using Illumina short-reads technology and genomicsequences were used to identify genetic variants. Variants were annotated integrating information from Ensembl100and literature knowledge on genes associated with embryonic development, miscarriages, lethality, cell cycle. Following annotation, variants were filtered to prioritize putatively detrimental variants in genes that are relevant for embryonic development using a pipeline that we developed. The code is available on gitHub (ezcn/grep). MAIN RESULTS AND THE ROLE OF CHANCE: Our pipeline prioritized 439 putatively causative single nucleotide polymorphisms among 11M variants discovered in ten embryos. By systematic investigation of all coding regions, 47 genes per embryo were selected. Among them STAG2, known in literature for its role in congenital and developmental disorders as well as in cancer,TLE4 a key gene in embryonic development, expressed in both embryonic and extraembryonic tissues in the Wnt and Notch signalling pathways, and FMNL2, involved in cell motility with a major role in driving cell migration. Our analysis is fully reproducible (our code is open-source), and we take measures to increase its robustness to false positives by excluding genes with>5% chance to be selected in a control population. LIMITATIONS, REASONS FOR CAUTION: This pilot study has major limitations in sample size and lack of integration of the parental genomic information. Despite being encouraging, the results need to be interpreted with caution as functional analyses are required to validate the hypotheses that have been generated. Although we have developed a robust and scalable methodology for prioritizing genetic variants, we have not yet extended it beyond the coding regions of the genome. WIDER IMPLICATIONS OF THE FINDINGS: This pilot study demonstrate that analysis of genome sequencing can help to clarify the causes of idiopathic miscarriages and provides initial results from the analysis of ten euploid embryos, discovering plausible candidate genes and variants. This study provides guidance for a larger study. Results of this and following wider studies can be used to test genetic predisposition to miscarriages in parents that are planning to conceive or undergoing preimplantation genetic testing. In a wider context, the results of this study might be relevant for genetic counseling and risk management in miscarriages.