This study introduces and validates a cloud condensation model on insoluble nuclei using a novel comparative analysis of simultaneous radiosonde and ceilometer data. A transformation of the radiosonde's temperature and relative humidity profiles into a simulated optical backscatter signal is implemented via the model, which includes a critical rate-limiting activation step. By comparing this simulated signal with the ceilometer's measured data, the model can determine the minimum effective size of dust particles required to act as cloud condensation nuclei. This approach has direct implications for improving the parameterization of cloud formation in local weather models, particularly for rainfall forecasting in dust-prone arid and semi-arid regions.

The recent spillover of highly pathogenic influenza A/H5N1 (HPAI-H5N1) viruses to cattle, other mammals, and humans poses a major risk to animal and human health. Virus adaptation to new species highlights the need for effective vaccines for animals and humans. We recently developed a rabies virus-based H5 vaccine encoding the HPAI-H5 antigen and presenting it on the surface of the rabies virus particle. To test the immunogenicity and efficacy of the vaccine in eliciting systemic and mucosal immune response, we vaccinated mice intramuscularly or intranasally with either live or inactivated and adjuvanted vaccine. The vaccine elicited neutralizing antibodies against RABV and H5N1 Influenza virus and protected mice from a lethal challenge with PR8 recombinants reassorted with the HA of clade 1 (Viet Nam 1203) or clade 2.3.4.4b HPAI-H5N1 viruses, highlighting its potential use in mitigating the risk of HPAI-H5N1 pandemic.

The remarkable success of mRNA vaccines in recent years has put the mRNA lipid nanoparticle (LNP) platform in the forefront of developing highly effective countermeasures against infectious diseases. Nevertheless, mRNA-LNPs have been studied for over three decades in the context of cancer therapy, and mRNA cancer vaccines are now being extensively pursued owing to recent advancements in this field. Although an anti-tumor effect mediated by mRNA-LNPs necessitates coordinated, multi-compartment, humoral and cellular immune responses, these crucial immune events have not been fully characterized.

Background: Botulism is a rare but potentially fatal neuroparalytic illness caused by Clostridium botulinum neurotoxins (BoNTs). While adult cases usually result from foodborne exposure or wound infection, intestinal colonization is exceedingly uncommon. Diagnosis can be delayed by overlap with other neuromuscular syndromes, and confirmation requires specialized assays. Anti-GQ1b antibodies, classically associated with Miller–Fisher syndrome (MFS), have rarely been reported in confirmed botulism, raising questions about shared pathophysiology. Case Presentation: We describe an adult patient with acute dyspnea, xerostomia, and cranial neuropathies. No foodborne source was identified, but intestinal colonization of BoNT/B toxigenic Clostridium botulinum was confirmed by stool enrichment and mouse lethality bioassay. The patient improved promptly following heptavalent antitoxin. Unexpectedly, anti-GQ1b antibodies were detected during recovery, a finding typically linked to MFS rather than botulism. Discussion: This case illustrates the diagnostic challenges of sporadic cases of botulism, especially when respiratory compromise and autonomic dysfunction dominate the initial presentation. The autoantibodies finding raises the possibility of molecular mimicry, whereby toxin–ganglioside interactions expose neuronal epitopes and trigger an immune response. While causality cannot be proven, the overlap between botulism and GQ1b-positive neuropathies merits further investigation. Conclusions: Clinicians should maintain high suspicion for botulism in adults with acute dyspnea, especially when associated with cranial neuropathies, even in the absence of foodborne exposure. Anti-ganglioside antibodies in this context should be interpreted with caution, as they do not exclude botulism but may highlight immunological overlap with autoimmune neuropathies.

The Bodélé Depression is the world's most intense dust source. One proposed explanation is wind amplification through an upwind mountain pass, the “wind-lens” effect, reflecting a Venturi-type acceleration. Using over 20 years of ERA5 reanalysis 10 m winds at the pass entrance and exit, together with Moderate Resolution Imaging Spectroradiometer Aerosol Optical Depth at the Bodélé, we show that winds nearly double as they traverse the pass. Two distinct dust-emission regimes emerge, separated by an exit-wind speed of ∼8 m/s. Under weaker winds, dust loading is low and largely independent of speed, while stronger winds produce sharply increasing dust and more aligned entrance–exit flow. These regimes coincide with seasonal shifts: weak, variable summer winds contrast with a stronger, directionally aligned winter regime when dust events are frequent. The strongest emissions occur when exit winds cluster near 48° and entrance winds have a northern component, reflecting optimal alignment with the corridor.

Background: The ability to test samples for the presence of specific tissues is useful for numerous forensic applications. More specifically, the identification of vital organ remains, such as the brain, in a crime scene or battlefield, can assist in determining the death of a missing person. In many cases, tissue samples are of insufficient quality or quantity for the application of histological methods, leaving forensic labs mostly restricted to immunological and catalytic assays designed to identify blood, semen, and saliva. Recent studies have suggested expression profiling-based methods for tissue and bodily fluid identification. Objective: We present a methylation-based assay for the detection of brain tissue in forensic samples. Methods: Genome-wide methylation data from 12 human tissues were analyzed to identify CpG sites uniquely methylated in brain tissue. Four candidate regions were selected based on high inter-tissue specificity and low intra-tissue variability. Targeted assays were developed using bisulfite conversion, polymerase chain reaction amplification, and next-generation sequencing, and validated based on reference tissues, mixtures, and environmentally degraded DNA samples. Results: Four regions displayed consistent brain-specific methylation with >94% single-read accuracy and complete sample-level discrimination at ≥5% brain DNA. The assay retained diagnostic performance in mixed and degraded samples, demonstrating robustness under typical forensic conditions. Conclusion: This study presents a sensitive and specific methylation-based assay for brain tissue identification. The approach enables reliable detection in degraded or composite materials and supports future integration of epigenetic biomarkers into forensic workflows for organ-source attribution.

Identifying emerging novel psychoactive substances (NPS), using liquid chromatography high-resolution tandem mass spectrometry (LC-HR-MS/MS), is often hindered by their frequent absence from mass spectral reference libraries. We developed a graph neural network (GNN) classification model to detect amphetamine derivatives, a subclass of NPS, based on their fragmentation patterns. The data set was generated by transforming high-resolution mass spectra from mass/charge (m/z) values into the chemical compositions of the fragments, from which fully connected graphs were constructed. Spectra were extracted from the NIST23 database, using only protonated HCD spectra to ensure consistency, and collision energy was included as a parameter to enhance spectral informativeness. The network architecture and hyperparameters were systematically optimized, including network layout and pooling layer type. To address the class imbalance in the data set, where amphetamine derivatives were outnumbered 100:1 by nonamphetamines, the loss function was adjusted to penalize false negatives 50 times more than false positives. Evaluation of model performance on a randomly selected 10% test subset yielded a recall of 0.86 and a precision of 0.11. The model successfully detected the spiked amphetamine derivative in a human plasma sample, achieving a 1% false positive rate and demonstrating its applicability to complex biological matrices.

Lipids, either as food components or as drug delivery formulations, are known to enhance the absorption of poorly water-soluble drugs in the gastrointestinal tract. Most research on the effects of different lipids and their lipolysis on dissolution and solubilization of poorly water-soluble drugs is conducted using liquid lipids, including liquid meals. In the present study, we examine the effects of a pure solid lipid, Tristearin (TS), compared to a liquid lipid, Triolein (TO), on the in vitro dissolution and the in vivo bioavailability of the BCS class II drug griseofulvin. Solid hydrophobic drugs may adsorb to solid lipids, and potentially decrease their bioavailability, however, we found that griseofulvin adsorbs negligibly to TS, which undergoes only partial lipolysis and therefore does not seem to contribute to griseofulvin dissolution nor to delay it and therefore TS is not expected to aid or impede griseofulvin bioavailability. Using the swine model to validate this prediction, we observed faster gastric emptying in the fasted state but similar overall bioavailabilities in the fasted and fed states, regardless of meal composition. This prompted an examination of the swine gastrointestinal contents for fed and fasted animals. Similar and high bile acids concentrations were found in fasted and fed states, suggesting a lack of difference in bile levels between the two states which may explain the observed lack of food effect. The use of the swine model to investigate lipid-enhanced bioavailability should be further explored, paying attention to the differences between different swine breeds. The results from in vitro experiments suggest that solid lipids are not expected to affect the bioavailability of BCS class II drugs.

Botulinum neurotoxin A (BoNT/A) is one of the most toxic substances known. The bacterium Clostridium botulinum produces it as a 150-kDa polypeptide with relatively low potency. Surprisingly, the endogenous protease that transforms this polypeptide into an activated, full-potency toxin, consisting of a 100-kDa heavy chain connected to a 50-kDa light chain (LC) by a disulfide bond, is still unknown. This study aimed to identify the BoNT/A-activating protease. We screened cation-exchange chromatography fractions of C. botulinum A culture supernatant for activity using a toxin-simulating substrate comprising the LC and the translocation domain (HN). Proteomic analysis of the active fraction identified α-clostripain as a candidate BoNT/A-activating protease. Recombinant α-clostripain cleaved the simulating substrate between the toxin LC and HN. However, incubation of recombinant α-clostripain with recombinant inactivated BoNT/A (rBoNT/Ai) resulted in non-specific digestion of the toxin. Since similar non-specific digestion was observed also by C. botulinum A culture supernatant, we hypothesized that the toxin should be protected by an accessory protein to prevent non-specific cleavage. Indeed, incubation of rBoNT/Ai with α-clostripain or culture supernatant in the presence of recombinant NTNH (non-toxic non-hemagglutinin) resulted in specific cleavage of the toxin into 100- and 50-kDa fragments. Subsequently, we evaluated the activation of rBoNT/A by α-clostripain in a mouse model. Cleavage of rBoNT/A by α-clostripain in the presence of NTNH resulted in a 77-fold increase in toxicity, corresponding to toxin activation. To the best of our knowledge, these results elucidate the mechanism of BoNT/A activation for the first time.

Air pollution interpolation is crucial for civil management. Various methods, including deterministic, geostatistical, and Machine Learning (ML)-based techniques have been utilized. Each method has limitations: deterministic methods do not adequately model environmental complexity, geostatistical methods struggle with small-scale areas, and ML depends heavily on data availability. On the other hand, air pollution simulators, which are driven by physicochemical dispersion models, capture intricate pollution dispersion patterns but are unsuitable for real-time interpolation.This study introduces an air pollution interpolation approach combining simulated air-dispersion patterns through an educated machine that includes machine learning and environmental modeling that considers boundaries, pollution sources, obstacles, wind dynamics, and topography. Specifically, we present a combined ML-based linear regression model designed to infer dense concentration maps from a sensor array using state-of-the-art simulation methods. We dub this the Ridiculously Simple data-driven air pollution Interpolation Method (RSIM). This method was evaluated on both synthetic and real-life-based simulation models.