In vitro oogenesis is key to elucidating the mechanism of human female germ‐cell development and its anomalies. Accordingly, pluripotent stem cells have been induced into primordial germ cell‐like cells and into oogonia with epigenetic reprogramming, yet further reconstitutions remain a challenge. Here, we demonstrate ex vivo reconstitution of fetal oocyte development in both humans and cynomolgus monkeys (Macaca fascicularis). With an optimized culture of fetal ovary reaggregates over three months, human and monkey oogonia enter and complete the first meiotic prophase to differentiate into diplotene oocytes that form primordial follicles, the source for oogenesis in adults. The cytological and transcriptomic progressions of fetal oocyte development in vitro closely recapitulate those in vivo. A comparison of single‐cell transcriptomes among humans, monkeys, and mice unravels primate‐specific and conserved programs driving fetal oocyte development, the former including a distinct transcriptomic transformation upon oogonia‐to‐oocyte transition and the latter including two active X chromosomes with little X‐chromosome upregulation. Our study provides a critical step forward for realizing human in vitro oogenesis and uncovers salient characteristics of fetal oocyte development in primates.

Single-cell RNA sequencing (scRNA-seq) can determine gene expression in numerous individual cells simultaneously, promoting progress in the biomedical sciences. However, scRNA-seq data are high-dimensional with substantial technical noise, including dropouts. During analysis of scRNA-seq data, such noise engenders a statistical problem known as the curse of dimensionality (COD). Based on high-dimensional statistics, we herein formulate a noise reduction method, RECODE (resolution of the curse of dimensionality), for high-dimensional data with random sampling noise. We show that RECODE consistently resolves COD in relevant scRNA-seq data with unique molecular identifiers. RECODE does not involve dimension reduction and recovers expression values for all genes, including lowly expressed genes, realizing precise delineation of cell fate transitions and identification of rare cells with all gene information. Compared with representative imputation methods, RECODE employs different principles and exhibits superior overall performance in cell-clustering, expression value recovery, and single-cell–level analysis. The RECODE algorithm is parameter-free, data-driven, deterministic, and high-speed, and its applicability can be predicted based on the variance normalization performance. We propose RECODE as a powerful strategy for preprocessing noisy high-dimensional data.

Human in vitro oogenesis provides a framework for clarifying the mechanism of human oogenesis. To create its benchmark, it is vital to promote in vitro oogenesis using a model physiologically close to humans. Here, we establish a foundation for in vitro oogenesis in cynomolgus (cy) monkeys (Macaca fascicularis): cy female embryonic stem cells harboring one active and one inactive X chromosome (Xa and Xi, respectively) differentiate robustly into primordial germ cell-like cells, which in xenogeneic reconstituted ovaries develop efficiently into oogonia and, remarkably, further into meiotic oocytes at the zygotene stage. This differentiation entails comprehensive epigenetic reprogramming, including Xi reprogramming, yet Xa and Xi remain epigenetically asymmetric with, as partly observed in vivo, incomplete Xi reactivation. In humans and monkeys, the Xi epigenome in pluripotent stem cells functions as an Xi-reprogramming determinant. We further show that developmental pathway over-activations with suboptimal upregulation of relevant meiotic genes impede in vitro meiotic progression. Cy in vitro oogenesis exhibits critical homology with the human system, including with respect to bottlenecks, providing a salient model for advancing human in vitro oogenesis.

No abstract

Single-cell RNA sequencing (scRNA-seq) can determine gene expression in numerous individual cells simultaneously, promoting progress in the biomedical sciences. However, scRNA-seq data are high-dimensional with substantial technical noise, including dropouts. During analysis of scRNA-seq data, such noise engenders a statistical problem known as the curse of dimensionality (COD). Based on high-dimensional statistics, we herein formulate a noise reduction method, RECODE (resolution of the curse of dimensionality), for high-dimensional data with random sampling noise. We show that RECODE consistently eliminates COD in relevant scRNA-seq data with unique molecular identifiers. RECODE does not involve dimension reduction and recovers expression values for all genes, including lowly expressed genes, realizing precise delineation of cell-fate transitions and identification of rare cells with all gene information. Compared to other representative imputation methods, RECODE employs different principles and exhibits superior overall performance in cell-clustering and single-cell level analysis. The RECODE algorithm is parameter-free, data-driven, deterministic, and high-speed, and notably, its applicability can be predicted based on the variance normalization performance. We propose RECODE as a general strategy for preprocessing noisy high-dimensional data.