Node injection for class-specific network poisoning.

Graph Neural Networks (GNNs) are powerful in learning rich network representations that aid the performance of downstream tasks. However, recent studies showed that GNNs are vulnerable to adversarial attacks involving node injection and network perturbation. Among these, node injection attacks are more practical as they do not require manipulation in the existing network and can be performed more realistically. In this paper, we propose a novel problem statement - a class-specific poison attack on graphs in which the attacker aims to misclassify specific nodes in the target class into a different class using node injection. Additionally, nodes are injected in such a way that they camouflage as benign nodes. We propose NICKI, a novel attacking strategy that utilizes an optimization-based approach to sabotage the performance of GNN-based node classifiers. NICKI works in two phases - it first learns the node representation and then generates the features and edges of the injected nodes. Extensive experiments and ablation studies on four benchmark networks show that NICKI is consistently better than four baseline attacking strategies for misclassifying nodes in the target class. We also show that the injected nodes are properly camouflaged as benign, thus making the poisoned graph indistinguishable from its clean version w.r.t various topological properties.

The bystander effect in rats.

To investigate whether the classic bystander effect is unique to humans, the effect of bystanders on rat helping was studied. In the presence of rats rendered incompetent to help through pharmacological treatment, rats were less likely to help due to a reduction in reinforcement rather than to a lack of initial interest. Only incompetent helpers of a strain familiar to the helper rat exerted a detrimental effect on helping; rats helped at near control levels in the presence of incompetent helpers from an unfamiliar strain. Duos and trios of potential helper rats helped at superadditive rates, demonstrating that rats act nonindependently with helping facilitated by the presence of competent-to-help bystanders. Furthermore, helping was facilitated in rats that had previously observed other rats' helping and were then tested individually. In sum, the influence of bystanders on helping behavior in rats features characteristics that closely resemble those observed in humans.

BrainFrame: a node-level heterogeneous accelerator platform for neuron simulations.

OBJECTIVE: The advent of high-performance computing (HPC) in recent years has led to its increasing use in brain studies through computational models. The scale and complexity of such models are constantly increasing, leading to challenging computational requirements. Even though modern HPC platforms can often deal with such challenges, the vast diversity of the modeling field does not permit for a homogeneous acceleration platform to effectively address the complete array of modeling requirements. APPROACH: In this paper we propose and build BrainFrame, a heterogeneous acceleration platform that incorporates three distinct acceleration technologies, an Intel Xeon-Phi CPU, a NVidia GP-GPU and a Maxeler Dataflow Engine. The PyNN software framework is also integrated into the platform. As a challenging proof of concept, we analyze the performance of BrainFrame on different experiment instances of a state-of-the-art neuron model, representing the inferior-olivary nucleus using a biophysically-meaningful, extended Hodgkin-Huxley representation. The model instances take into account not only the neuronal-network dimensions but also different network-connectivity densities, which can drastically affect the workload's performance characteristics. MAIN RESULTS: The combined use of different HPC technologies demonstrates that BrainFrame is better able to cope with the modeling diversity encountered in realistic experiments while at the same time running on significantly lower energy budgets. Our performance analysis clearly shows that the model directly affects performance and all three technologies are required to cope with all the model use cases. SIGNIFICANCE: The BrainFrame framework is designed to transparently configure and select the appropriate back-end accelerator technology for use per simulation run. The PyNN integration provides a familiar bridge to the vast number of models already available. Additionally, it gives a clear roadmap for extending the platform support beyond the proof of concept, with improved usability and directly useful features to the computational-neuroscience community, paving the way for wider adoption.

Unusually large erupted complex odontoma: A rare case report.

Odontomas are nonaggressive, hamartomatous developmental malformations composed of mature tooth substances and may be compound or complex depending on the extent of morphodifferentiation or on their resemblance to normal teeth. Among them, complex odontomas are relatively rare tumors. They are usually asymptomatic in nature. Occasionally, these tumors become large, causing bone expansion followed by facial asymmetry. Odontoma eruptions are uncommon, and thus far, very few cases of erupted complex odontomas have been reported in the literature. Here, we report the case of an unusually large, painless, complex odontoma located in the right posterior mandible.

1,2,4-thiadiazole: a novel Cathepsin B inhibitor.

A novel class of Cathepsin B inhibitors has been developed with a 1,2,4-thiadiazole heterocycle as the thiol trapping pharmacophore. Several compounds with different dipeptide recognition sequence (i.e., P1'-P2'=Leu-Pro-OH or P2-P1=Cbz-Phe-Ala) at the C5 position and with different substituents (i.e., OMe, Ph, or COOH) at the C3 position of the 1,2,4-thiadiazole ring have been synthesized and tested for their inhibitory activities. The substituted thiadiazoles 3a-h inhibit Cat B in a time dependent, irreversible manner. A mechanism based on active-site directed inactivation of the enzyme by disulfide bond formation between the active site cysteine thiol and the sulfur atom of the heterocycle is proposed. Compound 3a (K(i)=2.6 microM, k(i)K(i)=5630 M(-1)s(-1)) with a C3 methoxy moiety and a Leu-Pro-OH dipeptide recognition sequence, is found to be the most potent inhibitor in this series. The enhanced inhibitory potency of 3a is a consequence of its increased enzyme binding affinity (lower K(i)) rather than its increased intrinsic reactivity (higher k(i)). In addition, 3a is inactive against Cathepsin S, is a poor inhibitor of Cathepsin H and is >100-fold more selective for Cat B over papain.