Search for: All records

In the centrosymmetric title complexes, di-μ-acetato-bis({N,N-dimethyl-2-[phenyl(pyridin-2-yl)methylidene]hydrazine-1-carbothioamidato}zinc(II)), [Zn₂(C₁₅H₁₅N₄S)₂(C₂H₃O₂)₂] (I), and di-μ-acetato-bis({N-ethyl-2-[phenyl(pyridin-2-yl)methylidene]hydrazine-1-carbothioamidato}zinc(II)), [Zn₂(C₁₆H₁₇N₄S)₂(C₂H₃O₂)₂] (II), the zinc ions are chelated by theN,N,S-tridentate ligands and bridged by pairs of acetate ions. The acetate ion in (I) is disordered over two orientations in a 0.756 (6):0.244 (6) ratio, leading to different zinc coordination modes for the major (5-coordinate) and minor (6-coordinate) disorder components. Geometrical indices [τ₅= 0.32 and 0.30 for (I) (major component) and (II), respectively] suggest the zinc coordination in these phases to be distorted square pyramidal. This study forms part of our aim to discern the mechanism of metal binding in these chelators, their specificity and selectivity, and to gain insight into the role of cellular zinc in physiological processes such as infection, immunity and cancer. 

Wasserstein distances form a family of metrics on spaces of probability measures that have recently seen many applications. However, statistical analysis in these spaces is complex due to the nonlinearity of Wasserstein spaces. One potential solution to this problem is Linear Optimal Transport (LOT). This method allows one to find a Euclidean embedding, called {\it LOT embedding}, of measures in some Wasserstein spaces, but some information is lost in this embedding. So, to understand whether statistical analysis relying on LOT embeddings can make valid inferences about original data, it is helpful to quantify how well these embeddings describe that data. To answer this question, we present a decomposition of the {\it Fr\'echet variance} of a set of measures in the 2-Wasserstein space, which allows one to compute the percentage of variance explained by LOT embeddings of those measures. We then extend this decomposition to the Fused Gromov-Wasserstein setting. We also present several experiments that explore the relationship between the dimension of the LOT embedding, the percentage of variance explained by the embedding, and the classification accuracy of machine learning classifiers built on the embedded data. We use the MNIST handwritten digits dataset, IMDB-50000 dataset, and Diffusion Tensor MRI images for these experiments. Our results illustrate the effectiveness of low dimensional LOT embeddings in terms of the percentage of variance explained and the classification accuracy of models built on the embedded data. 

Disrupted iron balance causes anemia and iron overload leading to hypoxia and systemic oxidative stress. Iron overload may arise from red blood cell disorders such as sickle cell disease, thalassemia major and primary hemochromatosis, or from treatment with multiple transfusions. These hematological disorders are characterized by constant red blood cell hemolysis and the release of iron. Hemolysis is a continuous source of reactive oxygen species whose accumulation changes the redox potential in the erythrocyte, the endothelium and other tissue causing damage to organ systems. Iron overload and its consequences can be treated with iron chelating therapy. We have carried out structural studies of small molecule ligands that were previously reported for their iron chelating ability. The chelators were analyzed using mass spectrometry, proton nuclear magnetic resonance and infrared spectroscopy. The iron chelators, 2-benzoylpyridine-4,4-dimethyl-3-thiosemicarbazone, 3-ethyl-1-{[2-phenyl-1-(pyridin-2-yl)ethylidene]amino}thiourea and 1-{[2-phenyl-1-(pyridin-2-yl)ethylidene]amino}-3-(prop‑2-en-1-yl)thiourea in their unbound conformation were crystallized and their structures were determined. This work addresses the evolution of a thiosemicarbazone class of iron chelators by analyzing and comparing the structure and properties of a series of closely related molecules, relating these to their in vitro activity thus providing valuable update to the search for newer, better and more effective iron chelators and metal-based therapeutics. 

Neophobia, aversive behaviour towards novel objects, foods and environments, is a trait that affects the ability of animals to adapt to new environments and exploit novel resources. Our previous work demonstrated that individual responses of house sparrows (Passer domesticus) to object neophobia trials were not correlated with time spent in or latency to enter a novel environment. However, because no positive stimulus was present in the novel environment, this study may have measured spatialneophilia. In the present study, we placed familiar food dishes in a novel environment and assessed whether an individual’s willingness to enter and feed was significantly correlated with its willingness to feed from a familiar dish containing a novel object in the home cage. We exposed house sparrows (n= 26) to a novel environment and measured their latency to enter and feed, total time spent in the novel environment and total feeding time. Sparrows were also assessed for object neophobia in their home cage. Results indicated that there were no correlations between any of the measured behaviours in the novel environment and individual responses to novel object trials, suggesting that even with food as a common motivator, spatial neophobia and object neophobia represent two distinct traits. 

Archaeologists have recorded backed lithic specimens in the Pleistocene and Holocene around the world. One prominent hypothesis for the occurrence of backing is that it increases a stone tool's adhesion relative to what it would have otherwise been with unmodified, sharp edges. We conducted a highly controlled semi‐static tensile test in which we assessed lithic specimens that possessed both a backed and a non‐backed edge, opposing each other. We hafted each specimen's backed and non‐backed edges to wood, and the bi‐hafted stone implement was then pulled apart using a Universal Instron Materials Tester, allowing for a direct ‘head‐to‐head’ comparison of the two edge types’ adhesive properties. Our tensile test results suggested no significant difference between backed and non‐backed edges in terms of adhesion, which does not support the hypothesis that backing increases a lithic specimen's adhesion. 

Abstract Language models are typically evaluated on their success at predicting the distribution of specific words in specific contexts. Yet linguistic knowledge also encodes relationships between contexts, allowing inferences between word distributions. We investigate the degree to which pre-trained transformer-based large language models (LLMs) represent such relationships, focusing on the domain of argument structure. We find that LLMs perform well in generalizing the distribution of a novel noun argument between related contexts that were seen during pre-training (e.g., the active object and passive subject of the verb spray), succeeding by making use of the semantically organized structure of the embedding space for word embeddings. However, LLMs fail at generalizations between related contexts that have not been observed during pre-training, but which instantiate more abstract, but well-attested structural generalizations (e.g., between the active object and passive subject of an arbitrary verb). Instead, in this case, LLMs show a bias to generalize based on linear order. This finding points to a limitation with current models and points to a reason for which their training is data-intensive.1