Placental Multi-Omics Data-Mining in Intrauterine Growth Restriction

Background: Intrauterine Growth Restriction (IUGR) affects 8% of newborns and increases morbidity and mortality for the offspring even during later stages of life. Omics studies have evidenced epigenetic, gene expression and metabolic alterations in IUGR, but pathogenic mechanisms as a whole are not being fully understood and integrated together. An in-depth strategy combining multi-omics integration could help to better understand these pathogenic processes.   Methods: Methylomics and transcriptomics analyses were performed on 36 placenta samples in a case-control study. In addition to hierarchical clustering from the obtained quantitative data, machine learning algorithms were used to combine the analysis of more than 1200 genes found to be significantly modified. We used an automated text-mining approach, using the bulk textual gene annotations of the discriminant genes. The results were presented as clustered word clouds, allowing a rapid visualisation and summary of the pathophysiological processes involved. Support vector machine models were then used to explore the phenotypic subgroups (premature birth, birth weight and head circumference) associated with IUGR.   Findings: Gene annotation clustering highlighted the alteration of cell signalling and proliferation, cytoskeleton and intra- and extracellular structures, oxidative stress, protein turnover, muscle development, energy and lipid metabolism with insulin resistance. Models obtained using a support vector machine showed a high capacity for predicting the sub-phenotypes associated with IUGR.   Interpretation: Our results show that multi-omics integration, using the combination of gene annotation clustering, word cloud visualisation and support vector machine models, is highly relevant to provide an overview of the IUGR pathogenic mechanisms.   Registration Number: The study was validated by the French CPP (Comite de Protection des Personnes) and registered to the French Ministry of Research under number DC-2009-907. Funding Statement: This study was funded by a grant from the University Hospital of Angers, France. Declaration of Interests: The authors state: "none declared." Ethics Approval Statement: All placentas and blood samples were collected from Angers University Hospital. This study was approved by the Ethics Committee of Angers. All patients gave their written consent for the use of their placenta. Clinical data related to the mother and the foetus, as well as neonatal data, were collected from the patients’ obstetric files. The cohort was registered at the French CNIL (Commission Nationale de l’Informatique et des Libertes no. pWP03752UL, ethics committee for the collection of clinical data from patient records).