Cobaltabis(dicarbollide) ([o-COSAN]−) as Multifunctional Chemotherapeutics: A Prospective Application in Boron Neutron Capture Therapy (BNCT) for Glioblastoma

Purpose: The aim of our study was to assess if the sodium salt of cobaltabis(dicarbollide) and its di-iodinated derivative (Na[o-COSAN] and Na[8,8′-I2-o-COSAN]) could be promising agents for dual anti-cancer treatment (chemotherapy + BNCT) for GBM. Methods: The biological activities of the small molecules were evaluated in vitro with glioblastoma cells lines U87 and T98G in 2D and 3D cell models and in vivo in the small model animal Caenorhabditis elegans (C. elegans) at the L4-stage and using the eggs. Results: Our studies indicated that only spheroids from the U87 cell line have impaired growth after treatment with both compounds, suggesting an increased resistance from T98G spheroids, contrary to what was observed in the monolayer culture, which highlights the need to employ 3D models for future GBM studies. In vitro tests in U87 and T98G cells conclude that the amount of 10B inside the cells is enough for BNCT irradiation. BNCT becomes more effective on T98G after their incubation with Na[8,8′-I2-o-COSAN], whereas no apparent cell-killing effect was observed for untreated cells. Conclusions: These small molecules, particularly [8,8′-I2-o-COSAN]−, are serious candidates for BNCT now that the facilities of accelerator-based neutron sources are more accessible, providing an alternative treatment for resistant glioblastoma.

Side-by-Side Comparison of Five Chelators for 89Zr-Labeling of Biomolecules: Investigation of Chemical/Radiochemical Properties and Complex Stability

In this work, five different chelating agents, namely DFO, CTH-36, DFO*, 3,4,3-(LI-1,2-HOPO) and DOTA-GA, were compared with regard to the relative kinetic inertness of their corresponding 89Zr complexes to evaluate their potential for in vivo application and stable 89Zr complexation. The chelators were identically functionalized with tetrazines, enabling a fully comparable, efficient, chemoselective and biorthogonal conjugation chemistry for the modification of any complementarily derivatized biomolecules of interest. A small model peptide of clinical relevance (TCO-c(RGDfK)) was derivatized via iEDDA click reaction with the developed chelating agents (TCO = trans-cyclooctene and iEDDA = inverse electron demand Diels-Alder). The bioconjugates were labeled with 89Zr4+, and their radiochemical properties (labeling conditions and efficiency), logD(7.4), as well as the relative kinetic inertness of the formed complexes, were compared. Furthermore, density functional theory (DFT) calculations were conducted to identify potential influences of chelator modification on complex formation and geometry. The results of the DFT studies showed—apart from the DOTA-GA derivative—no significant influence of chelator backbone functionalization or the conjugation of the chelator tetrazines by iEDDA. All tetrazines could be efficiently introduced into c(RGDfK), demonstrating the high suitability of the agents for efficient and chemoselective bioconjugation. The DFO-, CTH-36- and DFO*-modified c(RGDfK) peptides showed a high radiolabeling efficiency under mild reaction conditions and complete 89Zr incorporation within 1 h, yielding the 89Zr-labeled analogs as homogenous products. In contrast, 3,4,3-(LI-1,2-HOPO)-c(RGDfK) required considerably prolonged reaction times of 5 h for complete radiometal incorporation and yielded several different 89Zr-labeled species. The labeling of the DOTA-GA-modified peptide was not successful at all. Compared to [89Zr]Zr-DFO-, [89Zr]Zr-CTH-36- and [89Zr]Zr-DFO*-c(RGDfK), the corresponding [89Zr]Zr-3,4,3-(LI-1,2-HOPO) peptide showed a strongly increased lipophilicity. Finally, the relative stability of the 89Zr complexes against the EDTA challenge was investigated. The [89Zr]Zr-DFO complex showed—as expected—a low kinetic inertness. Unexpectedly, also, the [89Zr]Zr-CTH-36 complex demonstrated a high susceptibility against the challenge, limiting the usefulness of CTH-36 for stable 89Zr complexation. Only the [89Zr]Zr-DFO* and the [89Zr]Zr-3,4,3-(LI-1,2-HOPO) complexes demonstrated a high inertness, qualifying them for further comparative in vivo investigation to determine the most appropriate alternative to DFO for clinical application.

TOrPEDO: witnessing model correctness with topological proofs

Model design is not a linear, one-shot process. It proceeds through refinements and revisions. To effectively support developers in generating model refinements and revisions, it is desirable to have some automated support to verify evolvable models. To address this problem, we recently proposed to adopt topological proofs, which are slices of the original model that witness property satisfaction. We implemented , a framework that provides automated support for using topological proofs during model design. Our results showed that topological proofs are significantly smaller than the original models, and that, in most of the cases, they allow the property to be re-verified by relying only on a simple syntactic check. However, our results also show that the procedure that computes topological proofs, which requires extracting unsatisfiable cores of LTL formulae, is computationally expensive. For this reason, currently handles models with a small dimension. With the intent of providing practical and efficient support for flexible model design and wider adoption of our framework, in this paper, we propose an enhanced—re-engineered—version of . The new version of relies on a novel procedure to extract topological proofs, which has so far represented the bottleneck of performances. We implemented our procedure within by considering Partial Kripke Structures (PKSs) and Linear-time Temporal Logic (LTL): two widely used formalisms to express models with uncertain parts and their properties. To extract topological proofs, the new version of converts the LTL formulae into an SMT instance and reuses an existing SMT solver (e.g., Microsoft ) to compute an unsatisfiable core. Then, the unsatisfiable core returned by the SMT solver is automatically processed to generate the topological proof. We evaluated by assessing (i) how does the size of the proofs generated by compares to the size of the models being analyzed; and (ii) how frequently the use of the topological proof returned by avoids re-executing the model checker. Our results show that provides proofs that are smaller ($$\approx $$ ≈ 60%) than their respective initial models effectively supporting designers in creating model revisions. In a significant number of cases ($$\approx $$ ≈ 79%), the topological proofs returned by enable assessing the property satisfaction without re-running the model checker. We evaluated our new version of by assessing (i) how it compares to the previous one; and (ii) how useful it is in supporting the evaluation of alternative design choices of (small) model instances in applied domains. The results show that the new version of is significantly more efficient than the previous one and can compute topological proofs for models with less than 40 states within two hours. The topological proofs and counterexamples provided by are useful to support the development of alternative design choices of (small) model instances in applied domains.

Ultrafast four-dimensional imaging of cardiac mechanical wave propagation with sparse optoacoustic sensing

Propagation of electromechanical waves in excitable heart muscles follows complex spatiotemporal patterns holding the key to understanding life-threatening arrhythmias and other cardiac conditions. Accurate volumetric mapping of cardiac wave propagation is currently hampered by fast heart motion, particularly in small model organisms. Here we demonstrate that ultrafast four-dimensional imaging of cardiac mechanical wave propagation in entire beating murine heart can be accomplished by sparse optoacoustic sensing with high contrast, ∼115-µm spatial and submillisecond temporal resolution. We extract accurate dispersion and phase velocity maps of the cardiac waves and reveal vortex-like patterns associated with mechanical phase singularities that occur during arrhythmic events induced via burst ventricular electric stimulation. The newly introduced cardiac mapping approach is a bold step toward deciphering the complex mechanisms underlying cardiac arrhythmias and enabling precise therapeutic interventions.

A New Knowledge Distillation Method for Object Detection Based on EMD

As a common method of model compression, the knowledge distillation method can distill the knowledge from the complex large model with strong learning ability to student small model with weak learning ability in the training process, to improve the accuracy and performance of the small model. At present, there has been much knowledge distillation methods specially designed for object detection and achieved good results. However, almost all methods failed to solve the problem of performance degradation caused by the high noise in the current detection framework. In this study, we proposed a feature automatic weight learning method based on EMD to solve these two problems. That is, the EMD method is used to process the space vector to reduce the impact of negative transfer and noise as much as possible, and at the same time, the weights are allocated adaptive to reduce student model’s learning from the teacher model with poor performance and make students more inclined to learn from good teachers. The loss (EMD Loss) was redesigned, and the HEAD was improved to fit our approach. We have carried out different comprehensive performance tests on multiple datasets, including PASCAL, KITTI, ILSVRC, and MS-COCO, and obtained encouraging results, which can not only be applied to the one-stage and two-stage detectors but also can be used radiatively with some other methods.

A Lightweight Modulation Classification Network Resisting White Box Gradient Attacks

Improving the attack resistance of the modulation classification model is an important means to improve the security of the physical layer of the Internet of Things (IoT). In this paper, a binary modulation classification defense network (BMCDN) was proposed, which has the advantages of small model scale and strong immunity to white box gradient attacks. Specifically, an end-to-end modulation signal recognition network that directly recognizes the form of the signal sequence is constructed, and its parameters are quantized to 1 bit to obtain the advantages of low memory usage and fast calculation speed. The gradient of the quantized parameter is directly transferred to the original parameter to realize the gradient concealment and achieve the effect of effectively defending against the white box gradient attack. Experimental results show that BMCDN obtains significant immune performance against white box gradient attacks while achieving a scale reduction of 6 times.

Fragment-based ab initio phasing of peptidic nanocrystals by MicroED

Microcrystal electron diffraction (MicroED) is transforming the visualization of molecules from nanocrystals, rendering their three-dimensional atomic structures from previously unamenable samples. Peptidic structures determined by MicroED include naturally occurring peptides, synthetic protein fragments and peptide-based natural products. However, as a diffraction method, MicroED is beholden to the phase problem, and its de novo determination of structures remains a challenge. ARCIMBOLDO, an automated, fragment-based approach to structure determination. It eliminates the need for atomic resolution, instead enforcing stereochemical constraints through libraries of small model fragments, and discerning congruent motifs in solution space to ensure validation. This approach expands the reach of MicroED to presently inaccessible peptidic structures including segments of human amyloids, and yeast and mammalian prions, and portends a more general phasing solution while limiting model bias for a wider set of chemical structures.

FACE

Cardinality estimation is one of the most important problems in query optimization. Recently, machine learning based techniques have been proposed to effectively estimate cardinality, which can be broadly classified into query-driven and data-driven approaches. Query-driven approaches learn a regression model from a query to its cardinality; while data-driven approaches learn a distribution of tuples, select some samples that satisfy a SQL query, and use the data distributions of these selected tuples to estimate the cardinality of the SQL query. As query-driven methods rely on training queries, the estimation quality is not reliable when there are no high-quality training queries; while data-driven methods have no such limitation and have high adaptivity. In this work, we focus on data-driven methods. A good data-driven model should achieve three optimization goals. First, the model needs to capture data dependencies between columns and support large domain sizes (achieving high accuracy). Second, the model should achieve high inference efficiency, because many data samples are needed to estimate the cardinality (achieving low inference latency). Third, the model should not be too large (achieving a small model size). However, existing data-driven methods cannot simultaneously optimize the three goals. To address the limitations, we propose a novel cardinality estimator FACE, which leverages the Normalizing Flow based model to learn a continuous joint distribution for relational data. FACE can transform a complex distribution over continuous random variables into a simple distribution (e.g., multivariate normal distribution), and use the probability density to estimate the cardinality. First, we design a dequantization method to make data more "continuous". Second, we propose encoding and indexing techniques to handle Like predicates for string data. Third, we propose a Monte Carlo method to efficiently estimate the cardinality. Experimental results show that our method significantly outperforms existing approaches in terms of estimation accuracy while keeping similar latency and model size.