Biology subjects

Bi, J.

Publications and source records attributed to Bi, J..

23 records · Page 2Linked to original sources

bioRxiv2020

Convolutional neural network for automated mass segmentation in mammography

BackgroundAutomatic segmentation and localization of lesions in mammogram (MG) images are challenging even with employing advanced methods such as deep learning (DL) methods. We developed a new model based on the architecture of the semantic segmentation U-Net model to precisely segment mass lesions in MG images. The proposed end-to-end convolutional neural network (CNN) based model extracts contextual information by combining low-level and high-level features. We trained the proposed model using huge publicly available databases, (CBIS-DDSM, BCDR-01, and INbreast), and a private database from the University of Connecticut Health Center (UCHC). ResultsWe compared the performance of the proposed model with those of the state-of-the-art DL models including the fully convolutional network (FCN), SegNet, Dilated-Net, original U-Net, and Faster R-CNN models and the conventional region growing (RG) method. The proposed Vanilla U-Net model outperforms the Faster R-CNN model significantly in terms of the runtime and the Intersection over Union metric (IOU). Training with digitized film-based and fully digitized MG images, the proposed Vanilla U-Net model achieves a mean test accuracy of 92.6%. The proposed model achieves a mean Dice coefficient index (DI) of 0.951 and a mean IOU of 0.909 that show how close the output segments are to the corresponding lesions in the ground truth maps. Data augmentation has been very effective in our experiments resulting in an increase in the mean DI and the mean IOU from 0.922 to 0.951 and 0.856 to 0.909, respectively. ConclusionsThe proposed Vanilla U-Net based model can be used for precise segmentation of masses in MG images. This is because the segmentation process incorporates more multi-scale spatial context, and captures more local and global context to predict a precise pixel-wise segmentation map of an input full MG image. These detected maps can help radiologists in differentiating benign and malignant lesions depend on the lesion shapes. We show that using transfer learning, introducing augmentation, and modifying the architecture of the original model results in better performance in terms of the mean accuracy, the mean DI, and the mean IOU in detecting mass lesion compared to the other DL and the conventional models.

bioinformatics↗

bioRxiv2020

Long-range gene regulation network of the MGMT enhancer in modulating glioma cell sensitivity to temozolomide

Acquired resistance to temozolomide (TMZ) is a major obstacle in glioblastoma treatment. MGMT, a DNA repair protein, and the methylation at its gene promoter, plays an important role in TMZ resistance. However, some evidences have suggested a MGMT-independent mechanisms underlying TMZ resistance. Here, we used MGMT enhancer as a model and discovered that its deletion in glioma cells of low MGMT expression induced increased sensitivity to temozolomide. Analysis of a combination of RNA-seq and Capture Hi-C further suggested multiple long-range target genes regulated by the MGMT enhancer and that interactions may play important roles in glioma cell sensitivity to TMZ. This study reveals a novel mechanism of regulation of TMZ sensitivity in glioma cells.

genetics↗

bioRxiv2020

CTCF-binding element regulates ESC differentiation via orchestrating long-range chromatin interaction between enhancers and HoxA

Proper expressions of Homeobox A cluster genes (HoxA) are essential for embryonic stem cells (ESC) differentiation and individual development. However, the mechanisms underlying controlling precise spatiotemporal expressions of HoxA during early ESC differentiation remain poorly understood. Herein, we identified a functional CTCF-binding element (CBE+47) closest to the 3-end of HoxA within the same topologically associated domain (TAD) in ESC. CRISPR-Cas9 mediated the CBE+47 deletion significantly promotes retinoic acid (RA)-induced HoxA expression and early ESC abnormal differentiation. In delineating its underlying mechanisms, we find that CBE+47 can precisely organize chromatin interactions between its adjacent enhancers and HoxA by using chromosome conformation capture assay (Capture-C). Furthermore, we also find that its adjacent enhancers as enhancer-enhancer interaction complex (EEIC) are required for RA-induced HoxA activation. Collectively, these results provide a new insight for RA-induced HoxA expression, also highlight the unique and precise regulatory roles of CBE in ESC differentiation.

developmental biology↗

bioRxiv2020

Integrin β3, a RACK1 interacting protein, is required for porcine reproductive and respiratory syndrome virus infection and NF-κB activation in Marc-145 cells

Porcine reproductive and respiratory syndrome virus (PRRSV) is the pathogen of porcine reproductive and respiratory syndrome (PRRS), which is one of the most economically harmful diseases in modern pig production worldwide. Receptor of activated protein C kinase 1 (RACK1) was previously shown to be indispensable for the PRRSV replication and NF-{kappa}B activation in Marc-145 cells. Here we identified a membrane protein, integrin {beta}3 (ITGB3), as a RACK1-interacting protein. PRRSV infection in Marc-145 cells upregulated the ITGB3 expression. Abrogation of ITGB3 by siRNA knockdown or antibody blocking inhibited PRRSV infection and NF-{kappa}B activation, while on the other hand, overexpression of ITGB3 enhanced PRRSV infection and NF-{kappa}B activation. Furthermore, inhibition of ITGB3 alleviated the cytopathic effects and reduced the TCID50 titer in Marc-145 cells. We also showed that RACK1 and ITGB3 were NF-{kappa}B target genes during PRRSV infection, and that they regulate each other. Our data indicate that ITGB3, presumably as a co-receptor, plays an imperative role for PRRSV infection and NF-{kappa}B activation in Marc-145 cells. PRRSV infection activates a positive feedback loop involving the activation of NF-{kappa}B and upregulation of ITGB3 and RACK1 in Marc-145 cells. The findings would advance our elaborated understanding of the molecular host-pathogen interaction mechanisms underlying PRRSV infection in swine and suggest ITGB3 and NF-{kappa}B signaling pathway as potential therapeutic targets for PRRS control. IMPORTANCEPorcine reproductive and respiratory syndrome virus (PRRSV) is one of the pathogens in pig production worldwide. Several cell surface receptors, such as heparan sulphate, sialoadhesin, vimentin and CD163, were identified to be involved in PRRSV infection in porcine alveolar macrophages (PAMs). We identified a cell surface protein, integrin {beta}3 (ITGB3), as an interacting protein with receptor of activated protein C kinase 1 (RACK1) from Marc-145 cells. ITGB3 interacts with RACK1 and facilitates PRRSV infection and NF-{kappa}B activation in Marc-145 cells, presumably as a co-receptor of CD136 or vimentin. Both ITGB3 and RACK1 were NF-{kappa}B target genes, and they regulate each other. The activation of NF-{kappa}B and the transcription of its downstream genes are beneficial for PRRSV infection/replication. The novel findings would advance our elaborated understanding of the molecular host-pathogen interaction mechanisms underlying PRRSV infection in swine and suggest ITGB3-RACK1-NF-{kappa}B axis as a potential therapeutic target for PRRS control.

microbiology↗

bioRxiv2019

A comparative assessment of aging-related NADPH diaphorase positivity in the spinal cord and medullary oblongata between pigeon and murine

NADPH diaphorase (N-d) positive neurons has been examined in many animals. N-d neurodegenerative neurites were detected in some animal models. However, detailed information of N-d positivity and aging related changes was still lack in the spinal cord and medulla oblongata of pigeons. In this study, we evaluated the N-d positivity and aging alterations in the spinal cord and medullary oblongata of the pigeon compared with rat and mouse. In pigeons, N-d neurons were more numerous in the dorsal horn, around the central canal and in the column of Terni in the thoracic and lumbar segments and scattered neurons occurred in the ventral horn of spinal segments. N-d neurons also occurred in the white matter of spinal cord. Morphometrical analysis demonstrated in the lumbosacral, cervical and thoracic regions. Compared with young pigeons, the size of N-d soma was significantly altered in aged pigeons. Meanwhile, the dramatic morphological changes occurred in the lumbar to sacral segments. The most important findings of this study were aging-related N-d positive bodies (ANB) in aged pigeons, mainly in the nucleus cuneatus externus (CuE), occasionally in the nuclei gracilis et cuneatus. ANBs were identified in the gracile nuclei in spinal cord in the aged rats and mice. ANBs were also detected in the CuE spinal nucleus in the aged rats. Immunohistochemistry also showed that the aging changes occurred in the cell types and neuropeptides in aged animals. The results suggested the weak inflammation and neuronal dysfunction in the spinal cord in aged pigeons. Our results suggested that the ANB could be considered as aging marker in the central nervous system.

neuroscience↗