Human pluripotent stem cells (hPSCs) have promising therapeutic applications due to their infinite capacity for self-renewal and pluripotency. Genomic stability is imperative for the clinical use of hPSCs; however, copy number variation (CNV), especially recurrent CNV at 20q11.21, may contribute genomic instability of hPSCs. Furthermore, the effects of CNVs in hPSCs at the whole-transcriptome scale are poorly understood. selleck products This study aimed to examine the functional in vivo and in vitro effects of frequently detected CNVs at 20q11.21 during early-stage differentiation of hPSCs. Comprehensive transcriptome profiling of abnormal hPSCs revealed that the differential gene expression patterns had a negative effect on differentiation potential. Transcriptional heterogeneity identified by single-cell RNA sequencing (scRNA-seq) of embryoid bodies from two different isogenic lines of hPSCs revealed alterations in differentiated cell distributions compared with that of normal cells. RNA-seq analysis of 22 teratomas identified several differentially expressed lineage-specific markers in hPSCs with CNVs, consistent with the histological results of the altered ecto/meso/endodermal ratio due to CNVs. Our results suggest that CNV amplification contributes to cell proliferation, apoptosis, and cell fate specification. This work shows the functional consequences of recurrent genetic abnormalities and thereby provides evidence to support the development of cell-based applications.Plasma concentrations of many cardiovascular and inflammatory proteins are altered after ST-elevation myocardial infarction (STEMI) and may provide prognostic information. We conducted a large-scale proteomic analysis in patients with STEMI, correlating protein levels to infarct size and left ventricular ejection fraction (LVEF) determined with cardiac magnetic resonance imaging. We analysed 131 cardiovascular and inflammatory proteins using a multiplex proximity extension assay and blood samples obtained at baseline, 6, 24, and 96 h from the randomised clinical trial CHILL-MI. Cardiac magnetic resonance imaging data at 4 ± 2 days and 6 months were available as per trial protocol. Using a linear regression model with bootstrap resampling and false discovery rate adjustment we identified five proteins (ST2, interleukin-6, pentraxin-3, interleukin-10, renin, and myoglobin) with elevated values corresponding to larger infarct size or worse LVEF and four proteins (TNF-related apoptosis-inducing ligand, TNF-related activation induced cytokine, interleukin-16, and cystatin B) with values inversely related to LVEF and infarct size, concluding that among 131 circulating inflammatory and cardiovascular proteins in the acute and sub-acute phase of STEMI, nine showed a relationship with infarct size and LVEF post-STEMI, with IL-6 and ST2 exhibiting the strongest association.To investigate the U2AF1 gene mutation site, mutation load and co-mutations genes in patients with myelodysplastic syndrome (MDS) and their effects on prognosis. Gene mutation detection by next-generation sequence and related clinical data of 234 MDS patients were retrospectively collected and analyzed for the relationship between the clinical characteristics, treatment efficacy and prognosis of U2AF1 gene mutation. Among the 234 MDS patients, the U2AF1 gene mutation rate was 21.7% (51 cases), and the median variant allele frequency was 39.5%. Compared with the wild type, the U2AF1 mutant had a higher incidence of chromosome 8 aberration, and was positively correlated with the occurrence of ASXL1, RUNX1, SETBP1 gene mutation, negatively correlated with SF3B1, NPM1 genes mutation (p  40% of U2AF1 is an independent factor of short OS in MDS patients. MDS patients with a mutation in the Q157P site of U2AF1 and a higher U2AF1 mutation load suggests poor prognosis, and co-mutated genes in U2AF1 can affect disease progression and prognosis.There is widespread evidence across Mars of past flows in major channel systems as well as more than one palaeo ocean level. However, evidence for the timing of channel flows and ocean levels is based on geographically diverse sources with a limited number of dates, making reconstructions of palaeo flows and ocean levels patchy. Here, based on high-resolution topography, image analysis and crater statistics, we have dated 35 different surfaces in Kasei Valles, that are predominantly found within erosional units enabling us to reconstruct a fascinating timeline of episodic flooding events (ranging from 3.7 to 3.6 Ga to ca. 2.0 Ga) interacting with changing ocean/base levels. The temporal correlation of the different surfaces indicates five periods of channel flows driving the evolution of Kasei Valles, in conjunction with the development of (at least) two ocean levels. Furthermore, our results imply that such ocean rose in elevation (ca. 1000 m) between ca. 3.6 Ga and 3.2 Ga and soon afterwards disappeared, thereby indicating a complex ancient Martian hydrosphere capable of supporting a vast ocean, with an active hydrological cycle stretching into the Amazonian.According to a recent study, around 99% of hospitals across the US now use electronic health record systems (EHRs). One of the most common types of EHR is the unstructured textual data, and unlocking hidden details from this data is critical for improving current medical practices and research endeavors. However, these textual data contain sensitive information, which could compromise our privacy. Therefore, medical textual data cannot be released publicly without undergoing any privacy-protective measures. De-identification is a process of detecting and removing all sensitive information present in EHRs, and it is a necessary step towards privacy-preserving EHR data sharing. Over the last decade, there have been several proposals to de-identify textual data using manual, rule-based, and machine learning methods. In this article, we propose new methods to de-identify textual data based on the self-attention mechanism and stacked Recurrent Neural Network. To the best of our knowledge, we are the first to employ these techniques. Experimental results on three different datasets show that our model performs better than all state-of-the-art mechanism irrespective of the dataset. Additionally, our proposed method is significantly faster than the existing techniques. Finally, we introduced three utility metrics to judge the quality of the de-identified data.