The automated classification of heart sounds plays a significant role in the diagnosis of cardiovascular diseases (CVDs). With the recent introduction of medical big data and artificial intelligence technology, there has been an increased focus on the development of deep learning approaches for heart sound classification. However, despite significant achievements in this field, there are still limitations due to insufficient data, inefficient training, and the unavailability of effective models. With the aim of improving the accuracy of heart sounds classification, an in-depth systematic review and an analysis of existing deep learning methods were performed in the present study, with an emphasis on the convolutional neural network (CNN) and recurrent neural network (RNN) methods developed over the last five years. This paper also discusses the challenges and expected future trends in the application of deep learning to heart sounds classification with the objective of providing an essential reference for further study.

Background: In studies of development and aging, the expression of many genes has been shown to undergo drastic changes at mRNA and protein levels. The connection between mRNA and protein expression level changes, as well as the role of posttranscriptional regulation in controlling expression level changes in postnatal development and aging, remains largely unexplored. Results: Here, we survey mRNA and protein expression changes in the prefrontal cortex of humans and rhesus macaques over developmental and aging intervals of both species’ lifespans. We find substantial decoupling of mRNA and protein expression levels in aging, but not in development. Genes showing increased mRNA/protein disparity in primate brain aging form expression patterns conserved between humans and macaques and are enriched in specific functions involving mammalian target of rapamycin (mTOR) signaling, mitochondrial function and neurodegeneration. Mechanistically, aging-dependent mRNA/protein expression decoupling could be linked to a specific set of RNA binding proteins and, to a lesser extent, to specific microRNAs. Conclusions: Increased decoupling of mRNA and protein expression profiles observed in human and macaque brain aging results in specific co-expression profiles composed of genes with shared functions and shared regulatory signals linked to specific posttranscriptional regulators. Genes targeted and predicted to be targeted by the aging-dependent posttranscriptional regulation are associated with biological processes known to play important roles in aging and lifespan extension. These results indicate the potential importance of posttranscriptional regulation in modulating aging-dependent changes in humans and other species.

Accurate prognosis of patients with cancer is important for the stratification of patients, the optimization of treatment strategies, and the design of clinical trials. Both clinical features and molecular data can be used for this purpose, for instance, to predict the survival of patients censored at specific time points. Multi-omics data, including genome-wide gene expression, methylation, protein expression, copy number alteration, and somatic mutation data, are becoming increasingly common in cancer studies. To harness the rich information in multi-omics data, we developed GDP (Group lass regularized Deep learning for cancer Prognosis), a computational tool for survival prediction using both clinical and multi-omics data. GDP integrated a deep learning framework and Cox proportional hazard model (CPH) together, and applied group lasso regularization to incorporate gene-level group prior knowledge into the model training process. We evaluated its performance in both simulated and real data from The Cancer Genome Atlas (TCGA) project. In simulated data, our results supported the importance of group prior information in the regularization of the model. Compared to the standard lasso regularization, we showed that group lasso achieved higher prediction accuracy when the group prior knowledge was provided. We also found that GDP performed better than CPH for complex survival data. Furthermore, analysis on real data demonstrated that GDP performed favorably against other methods in several cancers with large-scale omics data sets, such as glioblastoma multiforme, kidney renal clear cell carcinoma, and bladder urothelial carcinoma. In summary, we demonstrated that GDP is a powerful tool for prognosis of patients with cancer, especially when large-scale molecular features are available.

Spermatozoa acquire their fertilizing ability and forward motility during epididymal transit, suggesting the importance of the epididymis. Although the cell atlas of the epididymis was reported recently, the heterogeneity of the cells and the gene expression profile in the epididymal tube are still largely unknown. Considering single-cell RNA sequencing results, we thoroughly studied the cell composition, spatio-temporal differences in differentially expressed genes (DEGs) in epididymal segments and mitochondria throughout the epididymis with sufficient cell numbers. In total, 40,623 cells were detected and further clustered into 8 identified cell populations. Focused analyses revealed the subpopulations of principal cells, basal cells, clear/narrow cells, and halo/T cells. Notably, two subtypes of principal cells, the Prc7 and Prc8 subpopulations were enriched as stereocilia-like cells according to GO analysis. Further analysis demonstrated the spatially specific pattern of the DEGs in each cell cluster. Unexpectedly, the abundance of mitochondria and mitochondrial transcription (MT) was found to be higher in the corpus and cauda epididymis than in the caput epididymis by scRNA-seq, immunostaining, and qPCR validation. In addition, the spatio-temporal profile of the DEGs from the P42 and P56 epididymis, including transiting spermatozoa, was depicted. Overall, our study presented the single-cell transcriptome atlas of the mouse epididymis and revealed the novel distribution pattern of mitochondria and key genes that may be linked to sperm functionalities in the first wave and subsequent wave of sperm, providing a roadmap to be emulated in efforts to achieve sperm maturation regulation in the epididymis.

Alternative splicing (AS) has been implicated in cell cycle regulation and cancer, but the underlying mechanisms are poorly understood. The poly(U)-binding splicing factor 60 (PUF60) is essential for embryonic development and is overexpressed in multiple types of cancer. Here, we report that PUF60 promotes mitotic cell cycle and lung cancer progression by controlling AS of the cell division cycle 25C (CDC25C). Systematic analysis of splicing factors deregulated in lung adenocarcinoma (LUAD) identifies that elevated copy number and expression of PUF60 correlate with poor prognosis. PUF60 depletion inhibits LUAD cell-cycle G2/M transition, cell proliferation, and tumor development. Mechanistically, PUF60 knockdown leads to exon skipping enriched in mitotic cell cycle genes, including CDC25C. Exon 3 skipping in the full-length CDC25C results in nonsense-mediated mRNA decay and a decrease of CDC25C protein, thereby inhibiting cell proliferation. This study establishes PUF60 as a cell cycle regulator and an oncogenic splicing factor in lung cancer.

Spermatogenesis in testis is an important process for sexual reproduction, and worldwide about 10–15 percent of couples suffer from infertility. It is of importance to study spermatogenesis at single cell level in both of human and model organisms. Currently, single-cell RNA sequencing technologies (scRNA-seq) had been extensively applied to the study of cellular components and its gene regulations in the testes of different species, including human, monkey, mouse, and fly, but not in zebrafish. Zebrafish was a widely used model organism in biology and had been extensively used for the study of spermatogenesis in the previous studies. Therefore, it is also important to profile the transcriptome of zebrafish testis at single cell level. In this study, the transcriptomes of 14, 315 single cells from adult male zebrafish testes were profiled by scRNA-seq, and 10 cell populations were revealed, including Leydig cell, Sertoli cell, spermatogonia cell (SPG), spermatocyte, and spermatids. Notably, thousands of cell-type specific novel marker genes were identified, including sumo3b for SPG, krt18a.1 for Sertoli cells, larp1b and edrf1 for spermatids, which were also validated by RNA in situ hybridization experiments. Interestingly, through Ligand-Receptor (LR) analyses, zebrafish Leydig cells demonstrated stronger paracrine influence on germ cells than Sertoli cells. Overall, this study could be an important resource for the study of spermatogenesis in zebrafish and might also facilitate the study of the genes associated with human infertility through using zebrafish as a model organism.