Developing a molecular age model for oocyte assessment

The issue of advanced maternal age (AMA) has become a global issue, the average maternal age increased dramatically through the past decades. In vitro fertilization (IVF) is generally presented as a viable option but the success rate is limited by maternal age. The low success rate is also compound by subjective visual quality assessment, which creates a huge challenge for AMA women. A novel genetic‐based oocyte assessment platform for age prediction using mouse oocytes had been developed in this study. Based on the transcription profile of both immature (GV) and mature (MII) oocytes in young (6 weeks old) and aged group (12 months old) obtained by single cell RNA sequencing, we were able to identify the differential expression genes (DEGs) that govern the aging process. A total of 1377 DEGs in GV oocyte aging were identified. Of which 745 genes were up‐regulated and 631 genes were down‐regulated. On the other hand, we compared the transcriptome of MII oocytes and identified 1432 DEGs, 706 of them were upregulated and 726 were downregulated DEGs in aging process. The candidate genes for the age prediction model was selected by logistic regression model classifier with L1 norm penalty from the above sample, the power of prediction was ROC=0.81. Furthermore, we are developing a scoring system of 1–10 to give a clear grading regarding the oocyte quality. Based on the result obtained in mouse oocyte, a similar approach can be applied in human oocyte. Hence, the patients can have better idea about the oocyte quality and may proceed to oocyte treatments before IVF. Support or Funding Information Hong Kong Research Grant Council Collaborative Research Fund (C4054‐16G),