Indian River State College

Recent advancements in the field of cybersecurity has allowed for greater defensive technologies. The application of machine learning and artificial intelligence in cybersecurity has allowed for greater detection of cyber threats, especially in the field of network intrusion detection. Reinforcement learning, a prominent subfield in machine learning, aims to train an intelligent agent through providing rewards in response to stimuli in an environment. These advances, however, are often subject to adversarial attacks. In this study, we propose the application of a reinforcement learning-based network intrusion detection system by utilizing a deep recurrent Q-network for threat classification. We then implement adversarial machine learning attacks using the Fast Gradient Sign Method and Basic Iterative Method against our intrusion detection system. Our results indicate that the adversarial attacks were able to reduce the effectiveness of our intrusion detection systems on certain performance metrics by as much as 25%.

Zasmidium citri-griseum (syn. Mycosphaerella citri) is the identified causal agent of greasy spot/rind blotch disease in citrus, particularly from the Caribbean basin and other humid citrus production regions. The disease significantly impacts citrus health, causing greasy lesions on leaves, substantial defoliation, and rind blotch on fruit, leading to yield reductions and compromised marketability. In Florida, rind blotch has consistently posed challenges for grapefruit production, particularly affecting fresh fruit quality, and resulting in economic losses. This study presents the first draft genome of Z. citri-griseum, assembled using Illumina NovaSeq sequencing platform with a genome size of 44.3 Mb. Gene prediction identified 15,424 protein-coding genes, including 1,323 predicted secreted proteins and 481 pathogenicity-related effectors. This genomic resource provides insights into the biology, pathogenicity, evolutionary adaptations of Z. citri-griseum. It also supports the development of species-specific molecular tools and facilitates identification of lineage-specific adaptations within Mycosphaerellaceae (Mycosphaerellales, Dothideomycetes).

Delivery of biomolecules into plant vascular tissues remains a barrier to managing diseases caused by insect vector-borne pathogens and to modifying phenotypes of established perennial crops. Inspired by the vascularized growth of crown galls induced by Agrobacterium tumefaciens, we repurposed the bacterium’s plant growth regulator (PGR) genes to engineer autonomously dividing, transgene-expressing plant cell structures termed symbionts. A plant transformation vector (pSYM) incorporating the IaaM, IaaH, Ipt and gene5 cassette from A. tumefaciens strain C58 together with a gene of interest on the same transfer DNA was delivered to stems of herbaceous and woody dicots using disarmed A. tumefaciens strain EHA105. Symbiont morphology, vascular differentiation, transgene expression, molecular mobility and protein secretion were evaluated using microscopy, fluorescent reporters, dye tracing, RNA silencing assays and mass spectrometry-based proteomics. pSym inoculation reproducibly generated symbionts across diverse host plant species that were vascularly integrated into their host plants and transgene expression ranging from heterogeneous niches to more uniform patterns. Small molecules moved between symbionts and host vascular tissues, whereas larger proteins exhibited more restricted mobility. Post-transcriptional gene silencing signals moved freely throughout the symbiont and slightly into adjacent stem tissue. Under tested field and greenhouse conditions in potato and tomato, respectively, gall or symbiont formation had no negative impacts on plant growth or tuber and fruit yield. In vitro, symbiont cultures abundantly secreted recombinant protein into surrounding media. Together, these results establish symbionts as a modular, plant bioengineering platform capable of producing and potentially delivering biomolecules without modifying the host plant genome, providing a foundation for vascular-targeted therapeutics and phenotype modulation in crops.

Space and satellite-based systems have had a monumental impact on providing greater interconnectivity across the world. The usage of space and satellite-based systems has increased the ability to access internet resources even in remote areas. Unfortunately, these systems are subject to malicious and multi-faceted cyberattacks. Therefore, proper threat detection systems must be implemented to safeguard these space systems. In our study, we present our novel intrusion detection framework, SpIDER, a space satellite intrusion detection system using explainable reinforcement learning. SpIDER leverages the benefits offered by reinforcement learning and Shapley additive global explanations to improve both the performance and explainability of space-based intrusion detection. We compare our SpIDER framework to several popular machine learning algorithms using the STIN and NSL-KDD datasets. We observe that our SpIDER framework achieves high performance, with accuracy and G-Mean above 99.98% on the STIN satellite dataset. SpIDER also outperforms other machine learning models on the NSL-KDD local area network dataset, achieving accuracy of 76.71% and a G-Mean of 80.49%. These results demonstrate that our SpIDER explainable deep reinforcement learning framework can perform as well or better than supervised machine learning models on both satellite-style and local area network data.