FILM: Filtering and Machine Learning for Malware Detection in Edge Computing.

Machine learning with static-analysis features extracted from malware files has been adopted to detect malware variants, which is desirable for resource-constrained edge computing and Internet-of-Things devices with sensors; however, this learned model suffers from a misclassification problem because some malicious files have almost the same static-analysis features as benign ones. In this paper, we present a new detection method for edge computing that can utilize existing machine learning models to classify a suspicious file into either benign, malicious, or unpredictable categories while existing models make only a binary decision of either benign or malicious. The new method can utilize any existing deep learning models developed for malware detection after appending a simple sigmoid function to the models. When interpreting the sigmoid value during the testing phase, the new method determines if the model is confident about its prediction; therefore, the new method can take only the prediction of high accuracy, which reduces incorrect predictions on ambiguous static-analysis features. Through experiments on real malware datasets, we confirm that the new scheme significantly enhances the accuracy, precision, and recall of existing deep learning models. For example, the accuracy is enhanced from 0.96 to 0.99, while some files are classified as unpredictable that can be entrusted to the cloud for further dynamic or human analysis.

Evaluation of Graphene Oxide Induced Cellular Toxicity and Transcriptome Analysis in Human Embryonic Kidney Cells.

Graphene, a two-dimensional carbon sheet with single-atom thickness, shows immense promise in several nanoscientific and nanotechnological applications, including in sensors, catalysis, and biomedicine. Although several studies have shown the cytotoxicity of graphene oxide in different cell types, there are no comprehensive studies on human embryonic kidney (HEK293) cells that include transcriptomic analysis and an in vitro investigation into the mechanisms of cytotoxicity following exposure to graphene oxide. Therefore, we exposed HEK293 cells to different concentrations of graphene oxide for 24 h and performed several cellular assays. Cell viability and proliferation assays revealed a significant dose-dependent cytotoxic effect on HEK293 cells. Cytotoxicity assays showed increased lactate dehydrogenase (LDH) leakage and reactive oxygen species (ROS) generation, and decreased levels of reduced glutathione (GSH) and increased level of oxidized glutathione indicative of oxidative stress. This detailed mechanistic approach showed that graphene oxide exposure elicits significant decreases in mitochondrial membrane potential and ATP synthesis, as well as in DNA damage and caspase 3 activity. Furthermore, our RNA-Seq analysis revealed that HEK293 cells exposed to graphene oxide significantly altered the expression of genes involved in multiple apoptosis-related biological pathways. Moreover, graphene oxide exposure perturbed the expression of key transcription factors, promoting these apoptosis-related pathways by regulating their downstream genes. Our analysis provides mechanistic insights into how exposure to graphene oxide induces changes in cellular responses and massive cell death in HEK293 cells. To our knowledge, this is the first study describing a combination of cellular responses and transcriptome in HEK293 cells exposed to graphene oxide nanoparticles, providing a foundation for understanding the molecular mechanisms of graphene oxide-induced cytotoxicity and for the development of new therapeutic strategies.

Cytotoxicity and Transcriptomic Analyses of Biogenic Palladium Nanoparticles in Human Ovarian Cancer Cells (SKOV3).

Ovarian cancer incidence continues to increase at an alarming rate. Although various therapeutic approaches exist for ovarian cancer, they have limitations, including undesired side effects. Therefore, nanoparticle (NP)-mediated therapy may be a viable, biocompatible, and suitable alternative. To the best of our knowledge, no comprehensive analysis has been undertaken on the cytotoxicity and cellular pathways involved in ovarian cancer cells, particularly SKOV3 cells. Here, we investigated the effect of palladium NPs (PdNPs) and the molecular mechanisms and cellular pathways involved in ovarian cancer. We assayed cell viability, proliferation, cytotoxicity, oxidative stress, DNA damage, and apoptosis and performed an RNA-Seq analysis. The results showed that PdNPs elicited concentration-dependent decreases in cell viability and proliferation and induced increasing cytotoxicity at increasing concentrations, as determined by leakage of lactate dehydrogenase, increased levels of reactive oxygen species and malondialdehyde, and decreased levels of antioxidants like glutathione and superoxide dismutase. Furthermore, our study revealed that PdNPs induce mitochondrial dysfunction by altering mitochondrial membrane potential, reducing adenosine triphosphate levels, inducing DNA damage, and activating caspase 3, all of which significantly induced apoptosis in SKOV3 cells following PdNPs treatment. Gene ontology (GO) term analysis of PdNPs-exposed SKOV3 cells showed various dysregulated pathways, particularly nucleosome assembly, telomere organization, and rDNA chromatin silencing. When genes downregulated by PdNPs were applied to GO term enrichment analysis, nucleosome assembly was the top-ranked biological pathway. We also provide evidence for an association between PdNPs exposure and multiple layers of epigenetic transcriptional control and establish a molecular basis for NP-mediated apoptosis. These findings provide a foundation, potential targets, and novel insights into the mechanism underlying toxicity and pathways in SKOV3 cells, and open new avenues to identify novel targets for ovarian cancer treatment.