The Polycomb group protein EZH2 directly controls DNA methylation

The establishment and maintenance of epigenetic gene silencing is fundamental to cell determination and function. The essential epigenetic systems involved in heritable repression of gene activity are the Polycomb group (PcG) proteins and the DNA methylation systems. Here we show that the corresponding silencing pathways are mechanistically linked. We find that the PcG protein EZH2 (Enhancer of Zeste homolog 2) interacts-within the context of the Polycomb repressive complexes 2 and 3 (PRC2/3)-with DNA methyltransferases (DNMTs) and associates with DNMT activity in vivo. Chromatin immunoprecipitations indicate that binding of DNMTs to several EZH2-repressed genes depends on the presence of EZH2. Furthermore, we show by bisulphite genomic sequencing that EZH2 is required for DNA methylation of EZH2-target promoters. Our results suggest that EZH2 serves as a recruitment platform for DNA methyltransferases, thus highlighting a previously unrecognized direct connection between two key epigenetic repression systems.     

一代Technicolor模型对过程e~+e~-→tc的贡献

在ＬＥＰＩ的能量（Ｓ＝２００ＧｅＶ）下，我们计算了一代Ｔｅｃｈｎｉｃｏｌｏｒ（ＴＣ）模型对过程ｅ＋ｅ－→ｔｃ的贡献．计算结果表明，一代ＴＣ模型对ｅ＋ｅ－→ｔｃ的贡献很小，实验目前还无法观测     

Germline deletion of the miR-17∼92 cluster causes skeletal and growth defects in humans

MicroRNAs (miRNAs) are key regulators of gene expression in animals and plants. Studies in a variety of model organisms show that miRNAs modulate developmental processes. To our knowledge, the only hereditary condition known to be caused by a miRNA is a form of adult-onset non-syndromic deafness, and no miRNA mutation has yet been found to be responsible for any developmental defect in humans. Here we report the identification of germline hemizygous deletions of MIR17HG, encoding the miR-17～92 polycistronic miRNA cluster, in individuals with microcephaly, short stature and digital abnormalities. We demonstrate that haploinsufficiency of miR-17～92 is responsible for these developmental abnormalities by showing that mice harboring targeted deletion of the miR-17～92 cluster phenocopy several key features of the affected humans. These findings identify a regulatory function for miR-17～92 in growth and skeletal development and represent the first example of an miRNA gene responsible for a syndromic developmental defect in humans.     

Polyalanine expansion and frameshift mutations of the paired-like homeobox gene PHOX2B in congenital central hypoventilation syndrome

Congenital central hypoventilation syndrome (CCHS or Ondine's curse; OMIM 209880) is a life-threatening disorder involving an impaired ventilatory response to hypercarbia and hypoxemia. This core phenotype is associated with lower-penetrance anomalies of the autonomic nervous system (ANS) including Hirschsprung disease and tumors of neural-crest derivatives such as ganglioneuromas and neuroblastomas. In mice, the development of ANS reflex circuits is dependent on the paired-like homeobox gene Phox2b. Thus, we regarded its human ortholog, PHOX2B, as a candidate gene in CCHS. We found heterozygous de novo mutations in PHOX2B in 18 of 29 individuals with CCHS. Most mutations consisted of 5-9 alanine expansions within a 20-residue polyalanine tract probably resulting from non-homologous recombination. We show that PHOX2B is expressed in both the central and the peripheral ANS during human embryonic development. Our data support an essential role of PHOX2B in the normal patterning of the autonomous ventilation system and, more generally, of the ANS in humans.     

安络小皮伞(Marasmius androsaceus)Mun141菌株的选育

作者采用分离、纯化和诱变等方法，运用液体发酵，筛选出安络小皮伞（Ｍａｒｓｍｉｕｓａｎｄｒｏｓａｃｅｕｓ）Ｍｕｎ１４１高产菌株经３０ｔ发酵罐放大试验，结果显示，平均氯仿提取物含量为１２．３８％（标准差±４５．３８），较原菌株（ＭａＡ）提高５０．２４％，且性能稳定．同时，筛选出适宜Ｍｕｎ１４１生长的７＃固面培养基．Ｍｕｎ１４１菌株在７＃固体培养基上，菌丝萌发早，生长速度快，粗壮浓密，色泽纯正，其生长量高出原１＃固面培养基５－７％     

骨髓红粒细胞自动识别的深度学习模型

为了实现骨髓血细胞的自动识别, 构建了骨髓红系细胞和粒系细胞数据集, 基于深度学习语义分割技术提出了 CellNet 网络模型。 该模型通过加入残差模块增加了网络的深度, 利用卷积残差块使网络模型更容易训练, 并结合 U鄄Net 的裁剪操作为分割提供更精细的特征。 实验结果表明, 该模型对骨髓红系细胞和粒系细胞识别正确率分别达到 93. 65% 、95. 25% , 为骨髓血细胞自动识别技术提供了一种方法。     

基于改进ConvNeXt的COVID-19胸部图像分类

针对新型冠状病毒感染胸部 X-ray 图像分类任务数据集样本过少,现有的两阶段分类器和三阶段分类器模型对高纬度的图像特征提取效果差,模型训练慢等问题,提出一种基于 ConvNeXt 卷积神经网络改进的分类任务算法 ConvNeXt-AT。 ConvNeXt-AT 分类模型首先通过在 ConvNeXt Block 层添加混合域注意力机制 CBAM 来提高图像特征提取能力,不仅考虑了通道间的信息交互能力还考虑到了空间域上像素间的联系,得到 ConvNeXt-AT 模型;然后针对 X-ray 图片常见的泊松噪声使用全变差正则化方法对数据集进行降噪处理;最后在 COVID-19 公开的大型数据集共 21165 张图片进行对比实验。 实验结果表明,在训练数据集充分的情况下,改进的 ConvNeXt-AT 模型相较于常用分类模型 ResNet-50、MobileNet、EfficientNet 以及原 ConvNeXt-T 在准确率上分别提升了 2%、2. 7%、2. 1%、1. 9%。 最后通过 Grad-CAM 显示类激活图的图像可视化方法证明改进方法是可行的,模型具有很好的鲁棒性。