No abstract

The high level of sparsity in methylome profiles obtained using whole-genome bisulfite sequencing in the case of low biological material amount limits its value in the study of systems in which large samples are difficult to assemble, such as mammalian preimplantation embryonic development. The recently developed computational methods for addressing the sparsity by imputing missing have their limits when the required minimum data coverage or profiles of the same tissue in other modalities are not available. In this study, we explored the use of transfer learning together with Kullback-Leibler (KL) divergence to train predictive models for completing methylome profiles with very low coverage (below 2%). Transfer learning was used to leverage less sparse profiles that are typically available for different tissues for the same species, while KL divergence was employed to maximize the usage of information carried in the input data. A deep neural network was adopted to extract both DNA sequence and local methylation patterns for imputation. Our study of training models for completing methylome profiles of bovine oocytes and early embryos demonstrates the effectiveness of transfer learning and KL divergence, with individual increase of 29.98 and 29.43%, respectively, in prediction performance and 38.70% increase when the two were used together. The drastically increased data coverage (43.80–73.6%) after imputation powers downstream analyses involving methylomes that cannot be effectively done using the very low coverage profiles (0.06–1.47%) before imputation.

White adipose tissue plays an important role in energy storage. Excessive adiposity especially in the visceral adipose depot however has a stronger correlation with metabolic diseases such as insulin resistance. The specific anatomical locations of the visceral adipose tissue (VAT) suggest that it is subjective to depot‐specific regulation during development. Here, using a specific inducible lineage‐tracing mouse line, we identified that Tcf21 is specifically expressed in VAT but not in subcutaneous tissue. In VAT, Tcf21 is expressed in mesenchymal progenitor cells but not in differentiated adipocytes. Tcf21 lineage cells actively proliferate followed by differentiation into adipocytes during neonatal development but have a limited adipogenic capacity in adult mice even after high‐fat diet treatment. Bulk RNAseq and ATACseq analyses of Tcf21 lineage cells isolated from mice of different ages revealed the dynamic gene expression and chromatin accessibility in Tcf21 lineage cells. In particular, elevated expression of inflammatory genes and fibrotic genes were observed in Tcf21 lineage cells as the adiposity of mice increased. Using the transcriptomic and motif enrichment data, we predicted a gene regulatory network mediating the gene expression changes in Tcf21 lineage cells. Single‐cell RNAseq (scRNAseq) and immunostaining identified multiple subpopulations of Tcf21 lineage cells including 2 major subpopulations consisting of a mesothelial subpopulation and an interstitial subpopulation, as well as a small population that expressed select inflammatory genes exclusively in obese mice. Using an inducible cell‐type‐specific Tcf21 knockout mouse line, we identified that neonatal deletion of Tcf21 in mice led to increased adipogenesis of Tcf21 lineage cells during postnatal development and improved metabolism after high‐fat diet treatment. In vitro loss‐of‐function and gain‐of‐function studies showed that Tcf21 inhibits the adipogenic differentiation of VAT progenitor cells. Bulk RNAseq and scRNAseq showed that Tcf21 lineage cells from Tcf21 knockout mice were developmentally in advance of those from their WT littermates. Mechanistic studies identified that Tcf21 inhibits adipogenesis through promoting the expression of Dlk1, a known negative regulator of adipogenesis, in the interstitial subpopulation of Tcf21 lineage cells.

Multi-Task Learning (MTL) is a powerful technique that has gained popularity due to its performance improvement over traditional Single-Task Learning (STL). However, MTL is often challenging because there is an exponential number of possible task groupings, which can make it difficult to choose the best one, and some groupings might produce performance degradation due to negative interference between tasks. Furthermore, existing solutions are severely suffering from scalability issues, limiting any practical application. In our paper, we propose a new data-driven method that addresses these challenges and provides a scalable and modular solution for classification task grouping based on hand-crafted features, specifically Data Maps, which capture the training behavior for each classification task during the MTL training. We experiment with the method demonstrating its effectiveness, even on an unprecedented number of tasks (up to 100).