Search

This study investigated KLF7’s effect on sugar induced retinal ganglion cells (RGCs) biological activity. The RGCs cells divided into blank group (RA), high sugar group (RB), high sugar+NC group (RC) and high sugar+KLF7 group (RD) (transfected with KLF7 mimic) followed by analysis cell proliferation by MTT, cell apoptosis by flow cytometry and protein expression by western blot and ROS level. RB and RC group showed significantly reduced KLF7 mRNA and protein level compared to RA group (P < 0.05) without different between RB and RC group (P > 0.05). RD group had significantly increased LKF7 and Sirt1 protein expression (F = 113.3, P < 0.0, 01), reduced cell proliferation (P < 0.05) and increased RGCs apoptosis rate (P < 0.05) compared with RB and RC group. After 24 h, RB and RC group presented significantly higher ROS level (P < 0.05) which was reduced in RD group (P < 0.05). In conclusion, KLF7 can change sugar induced retinal ganglion cell biological activity and reduce the oxidative stress level.

We aimed to uncover the role of microRNA-181 (miR-181) in the disease onset of diabetic retinopathy (DR) and its underlying mechanism.MiR-181 levels in plasma and aqueous humor samples of non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR) and healthy subjects were analyzed by microarray and quantitative real-time polymerase chain reaction (qRT-PCR). Proliferative and migrative capacities of human retinal endothelial cells (hRECs) regulated by miR-181 were assessed. The binding between miR-181 and Kruppel-like factor 6 (KLF6) was verified by dual-luciferase reporter assay.MiR-181 was upregulated in plasma and aqueous humor samples of NPDR and PDR patients. Overexpression of miR-181 stimulated hRECs to proliferate and migrate. KLF6 was the downstream gene binding miR-181, which was involved in the regulation of hRECs by miR-181.MiR-181 is upregulated in plasma and aqueous humor of DR patients. It enhances proliferative and migratory potentials of retinal endothelial cells by targeting KLF6.

Brain tumor segmentation methods using deep neural networks have recently achieved significant performance breakthroughs. However, the existing brain tumor segmentation networks are directly implemented on whole brain images, resulting in possibly reduced segmentation performance due to the disturbance of background regions. To solve this problem, inspired by the Mask R-CNN, a novel brain tumor segmentation model called BrainSeg R-CNN is proposed in this work, which classifies the brain tumor areas and boundaries based on the detected region of interest in an end-to-end manner to achieve segmentation result. Also, an effective feature extraction strategy is presented in BrainSeg R-CNN, which in detail extracts various kinds of information from separate channels for each modality and immediately adopts a cross-connection operator to realize the information transmission among different channels. Moreover, concatenation and add calculation are integrated to improve the fusion efficiency of multi-scale features from brain tumor images. Additionally, a multi-weighted and multi-task loss function which fully considers tumor size and overlap label is introduced, significantly improving the segmentation performance. Experimental results on BraTS 2017 dataset demonstrate that our BrainSeg R-CNN obtains competitive performance with state-of-the-arts.

The Chinese chestnut (Castanea mollissima Bl.) is a woody nut crop with a high ecological value. Although many cultivars have been selected from natural seedlings, elite lines with comprehensive agronomic traits and characters remain rare. To explore genetic resources with aid of whole genome sequence will play important roles in modern breeding programs for chestnut. In this study, we generated a high-quality C. mollissima genome assembly by combining 90× Pacific Biosciences long read and 170× high-throughput chromosome conformation capture data. The assembly was 688.93 Mb in total, with a contig N50 of 2.83 Mb. Most of the assembled sequences (99.75%) were anchored onto 12 chromosomes, and 97.07% of the assemblies were accurately anchored and oriented. A total of 33,638 protein-coding genes were predicted in the C. mollissima genome. Comparative genomic and transcriptomic analyses provided insights into the genes expressed in specific tissues, as well as those associated with burr development in the Chinese chestnut. This highly contiguous assembly of the C. mollissima genome provides a valuable resource for studies aiming at identifying and characterizing agronomical-important traits, and will aid the design of breeding strategies to develop more focused, faster, and predictable improvement programs.

Segmentation is a routine and crucial procedure for the treatment of brain tumors. Deep learning based brain tumor segmentation methods have achieved promising performance in recent years. However, to pursue high segmentation accuracy, most of them require too much memory and computation resources. Motivated by a recently proposed partially reversible U-Net architecture that pays more attention to memory footprint, we further present a novel Memory-Efficient Cascade 3D U-Net (MECU-Net) for brain tumor segmentation in this work, which can achieve comparable segmentation accuracy with less memory and computation consumption. More specifically, MECU-Net utilizes fewer down-sampling channels to reduce the utilization of memory and computation resources. To make up the accuracy loss, MECU-Net employs multi-scale feature fusion module to enhance the feature representation capability. Additionally, a light-weight cascade model, which resolves the problem of small target segmentation accuracy caused by model compression to some extent, is further introduced into the segmentation network. Finally, edge loss and weighted dice loss are combined to refine the brain tumor segmentation results. Experiment results on BraTS 2019 validation set illuminate that MECU-Net can achieve average Dice coefficients of 0.902, 0.824 and 0.777 on the whole tumor, tumor core and enhancing tumor, respectively.