Quercetin is an organic flavonoid present in several fruits and vegetables. The anti-inflammatory, antiviral, antioxidant, cardio-protective, anti-carcinogenic and neuroprotective properties demonstrated by this dietary supplement endorses it as a possible treatment for inflammatory diseases and cancer. Unfortunately, conflicting research has cast uncertainties on the toxicity of quercetin. The main purpose of this study was to determine if quercetin has any toxic properties in mice at doses that have shown efficacy in pre-clinical studies regarding cancer, cancer therapy, and their off-target effects.A sub-chronic toxicity study of quercetin was examined in male and female CD2F1 mice. Three different doses of quercetin (62, 125, and 250 mg/kg of diet) were infused into the AIN-76A purified diet and administered to mice ad libitum for 98 days. Body weight (BW), food consumption, water intake, body composition, blood count, behavior, and metabolic phenotype were assessed at various timepoints during the course of the experiment. Tissue and organs were evaluated for gross pathological changes and plasma was used to measure alkaline phosphatase (AP), aspartate transaminase (AST), and alanine transaminase (ALT).We found that low (62 mg/kg of diet), medium (125 mg/kg of diet), and high (250 mg/kg of diet) quercetin feeding had no discernible effect on body composition, organ function, behavior or metabolism.In summary, our study establishes that quercetin is safe for use in both female and male CD2F1 mice when given at ~ 12.5, 25, or 50 mg/kg of BW daily doses for 14 weeks (i.e. 98 days). Further studies will need to be conducted to determine any potential toxicity of quercetin following chronic ingestion.
Human induced pluripotent stem (hiPS) cells have the ability to undergo self-renewal and differentiation similarly to human embryonic stem (hES) cells. We have recently shown that hES cells under replication stress fail to activate checkpoint kinase 1 (CHK1). They instead commit to apoptosis, which appears to be a primary defense mechanism against genomic instability. It is not known whether the failure of CHK1 activation and activation of apoptosis under replication stress is solely a feature of hES cells, or if it is a feature that can be extended to hiPS cells.Here we generated integration-free hiPS cell lines by mRNA transfection, and characterised the cell lines. To investigate the mechanism of S phase checkpoint activation, we have induced replication stress by adding excess thymidine to the cell culture medium, and performed DNA content analysis, apoptosis assays and immunoblottings.We are showing that hiPS cells similarly to hES cells, fail to activate CHK1 when exposed to DNA replication inhibitors and commit to apoptosis instead. Our findings also suggest the Ataxia Telangiectasia Mutated pathway might be responding to DNA replication stress, resulting in apoptosis.Together, these data suggest that the apoptotic response was properly restored during reprogramming with mRNA, and that apoptosis is an important mechanism shared by hiPS and hES cells to maintain their genomic integrity when a replication stress occurs.
Deploying convolutional neural networks (CNNs) for embedded applications presents many challenges in balancing resource-efficiency and task-related accuracy. These two aspects have been well-researched in the field of CNN compression. In real-world applications, a third important aspect comes into play, namely the robustness of the CNN. In this paper, we thoroughly study the robustness of uncompressed, distilled, pruned and binarized neural networks against white-box and black-box adversarial attacks (FGSM, PGD, C&W, DeepFool, LocalSearch and GenAttack). These new insights facilitate defensive training schemes or reactive filtering methods, where the attack is detected and the input is discarded and/or cleaned. Experimental results are shown for distilled CNNs, agent-based state-of-the-art pruned models, and binarized neural networks (BNNs) such as XNOR-Net and ABC-Net, trained on CIFAR-10 and ImageNet datasets. We present evaluation methods to simplify the comparison between CNNs under different attack schemes using loss/accuracy levels, stress-strain graphs, box-plots and class activation mapping (CAM). Our analysis reveals susceptible behavior of uncompressed and pruned CNNs against all kinds of attacks. The distilled models exhibit their strength against all white box attacks with an exception of C&W. Furthermore, binary neural networks exhibit resilient behavior compared to their baselines and other compressed variants.
Closing the gap between the hardware requirements of state-of-the-art convolutional neural networks and the limited resources constraining embedded applications is the next big challenge in deep learning research. The computational complexity and memory footprint of such neural networks are typically daunting for deployment in resource constrained environments. Model compression techniques, such as pruning, are emphasized among other optimization methods for solving this problem. Most existing techniques require domain expertise or result in irregular sparse representations, which increase the burden of deploying deep learning applications on embedded hardware accelerators. In this paper, we propose the autoencoder-based low-rank filter-sharing technique technique (ALF). When applied to various networks, ALF is compared to state-of-the-art pruning methods, demonstrating its efficient compression capabilities on theoretical metrics as well as on an accurate, deterministic hardware-model. In our experiments, ALF showed a reduction of 70\% in network parameters, 61\% in operations and 41\% in execution time, with minimal loss in accuracy.
Dense pixel matching is important for many computer vision tasks such as disparity and flow estimation. We present a robust, unified descriptor network that considers a large context region with high spatial variance. Our network has a very large receptive field and avoids striding layers to maintain spatial resolution. These properties are achieved by creating a novel neural network layer that consists of multiple, parallel, stacked dilated convolutions (SDC). Several of these layers are combined to form our SDC descriptor network. In our experiments, we show that our SDC features outperform state-of-the-art feature descriptors in terms of accuracy and robustness. In addition, we demonstrate the superior performance of SDC in state-of-the-art stereo matching, optical flow and scene flow algorithms on several famous public benchmarks.
Defects in neural crest development have been implicated in many human disorders, but information about human neural crest formation mostly depends on extrapolation from model organisms. Human pluripotent stem cells (hPSCs) can be differentiated into in vitro counterparts of the neural crest, and some of the signals known to induce neural crest formation in vivo are required during this process. However, the protocols in current use tend to produce variable results, and there is no consensus as to the precise signals required for optimal neural crest differentiation. Using a fully defined culture system, we have now found that the efficient differentiation of hPSCs to neural crest depends on precise levels of BMP signaling, which are vulnerable to fluctuations in endogenous BMP production. We present a method that controls for this phenomenon and could be applied to other systems where endogenous signaling can also affect the outcome of differentiation protocols.
Cachexia, a complex wasting syndrome, significantly affects the quality of life and treatment options for cancer patients. Studies have reported a strong correlation between high platelet count and decreased survival in cachectic individuals. Therefore, this study aimed to investigate the immunopathogenesis of cancer cachexia using the Apc
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
