General Graph Neural Network-Based Model To Accurately Predict Cocrystal Density and Insight from Data Quality and Feature Representation.

Cocrystal engineering as an effective way to modify solid-state properties has inspired great interest from diverse material fields while cocrystal density is an important property closely correlated with the material function. In order to accurately predict the cocrystal density, we develop a graph neural network (GNN)-based deep learning framework by considering three key factors of machine learning (data quality, feature presentation, and model architecture). The result shows that different stoichiometric ratios of molecules in cocrystals can significantly influence the prediction performances, highlighting the importance of data quality. In addition, the feature complementary is not suitable for augmenting the molecular graph representation in the cocrystal density prediction, suggesting that the complementary strategy needs to consider whether extra features can sufficiently supplement the lacked information in the original representation. Based on these results, 4144 cocrystals with 1:1 stoichiometry ratio are selected as the dataset, supplemented by the data augmentation of exchanging a pair of coformers. The molecular graph is determined to learn feature representation to train the GNN-based model. Global attention is introduced to further optimize the feature space and identify important atoms to realize the interpretability of the model. Benefited from the advantages, our model significantly outperforms three competitive models and exhibits high prediction accuracy for unseen cocrystals, showcasing its robustness and generality. Overall, our work not only provides a general cocrystal density prediction tool for experimental investigations but also provides useful guidelines for the machine learning application. All source codes are freely available at https://github.com/Xiao-Gua00/CCPGraph.

Synergistic Procoagulant Mechanism and Application of Kaolin-Zeolite Composite Hemostat for Effective Hemorrhage Control.

Achieving timely and effective hemorrhage control is imperative for the survival of individuals with severe bleeding. Hemostatic materials, by enhancing the natural cell-based coagulation response, are essential tools in modern and military medical practice for controlling bleeding, especially in emergency and surgical settings. Here, we report a new type of composite hemostatic material with two different aluminosilicate-based components, kaolin and zeolite, which synergistically work together in different stages of the coagulation cascade reactions. Kaolin can effectively activate the clotting factor FXII in the early stage, and zeolite can accumulate and assemble FXa and FVa on its surface and thereafter lead to the formation of highly active thrombin in the later stage. The synergistic action mechanism between kaolin and zeolite significantly boosts the levels of FXIIa and FXa, and it also greatly enhances plateau thrombin activity. For practical application, a kaolin-modified zeolite gauze is fabricated, and it demonstrates excellent hemostatic effectiveness. Compared to the combat gauze currently used in front-line treatment, it reduces blood loss by 75% and shortens hemostasis time by 33% in a rabbit femoral artery injury model. In addition, this kaolin-zeolite gauze has no heat release problem and a nearly zero particle shedding rate, which greatly decreases the safety risk compared to current commercial inorganic-based hemostatic gauzes.

Paraffin-Coated Hydrophobic Hemostatic Zeolite Gauze for Rapid Coagulation with Minimal Adhesion.

To solve the problem of strong adhesion and excessive blood loss caused by the use of hydrophilic zeolite gauze (Z-Gauze) in uncontrollable bleeding, we have modified the surface of commercial Z-Gauze with a paraffin coating and prepared a hydrophobic dressing PZ-Gauze. After paraffin coating, the adhesion of Z-Gauze was reduced without an obvious decrease in coagulation activity. The clotting time of the hydrophobic PZ-Gauze was reduced from 378.3 to 154.6 s compared with that of cotton gauze, and the peeling force was decreased from 348.8 to 84.7 mN compared with that of Z-Gauze. Besides, PZ-Gauze can efficiently cut down the blood loss during treatment. On the basis of in vitro and in vivo experiments, it is confirmed that surface hydrophobic modification does not change the procoagulant performance because of the maintained cation exchange capacity of zeolites, and the reduced blood loss as well as enhanced difficulty for fibrin adhesion is attributed to its hydrophobicity. This is different from the traditional gauze procoagulant theories, where gauze hydrophilicity and procoagulant performance are always positively correlated.

Susceptibility-diffusion mismatch correlated with leptomeningeal collateralization in large vessel occlusion stroke.

OBJECTIVE: To investigate the relationship between asymmetric prominent hypointense vessels (prominent vessel sign, PVS) on susceptibility-weighted imaging (SWI) and leptomeningeal collateralization in patients with acute ischemic stroke due to large vessel occlusion. METHODS: We retrospectively enrolled patients with M1 segment occlusion of the middle cerebral artery who underwent emergency magnetic resonance imaging and digital subtraction angiography within 24 hours from stroke onset. The extent of PVS on SWI was assessed using the Alberta Stroke Program Early CT Score (ASPECTS). Leptomeningeal collateralization on digital subtraction angiography images was assessed using the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology (ASITN/SIR) scale. Spearman's rank correlation test was performed to explore the correlation of ASITN/SIR scores with SWI-ASPECTS and SWI-diffusion-weighted imaging (DWI) mismatch scores. RESULTS: Thirty-five patients were enrolled. There was no significant correlation between SWI-ASPECTS and ASITN/SIR scores. However, SWI-DWI mismatch scores were positively correlated with ASITN/SIR scores. CONCLUSION: The range of PVS on SWI did not closely reflect the collateral status, while the range of SWI-DWI mismatch was significantly correlated with the leptomeningeal collateralization. In patients with acute anterior circulation stroke due to large vessel occlusion, larger SWI-DWI mismatch was associated with better leptomeningeal collaterals.

Neuroprotective Influence of miR-301a Inhibition in Experimental Cerebral Ischemia/Reperfusion Rat Models Through Targeting NDRG2.

The objective of this study is to find out the potential influence of miR-301a in an experimental cerebral ischemia-reperfusion (I/R) rat model through targeting NDRG2. Rats with cerebral I/R injury were constructed and classified into model, miR-301a inhibitor, miR-301a mimic, NC (negative control), siNDRG2, NDRG2, and miR-301a inhibitor + si-NDRG2 groups, as well as another sham group. Cerebral infarct volume and cell apoptosis were observed by TTC staining and TUNEL staining. The targeting relationship between miR-301a and NDRG2 was verified by luciferase assay. ELISA, qRT-PCR, and Western blot were used to detect the expressions of related molecules. Compared with sham group, rats in the model group had elevated neurological function score and infarct volume; meanwhile, the cell apoptosis rate and inflammatory response were also increased with enhanced expression of miR-301a and NDRG2 (all P < 0.05). These changes were worsened in the miR-301a mimic and si-NDRG2 groups. Conversely, those rats in the miR-301a inhibitor and NDRG2 groups presented increased NDRG2, and at the same time, other above concerning factors also exhibited opposite tendencies (all P < 0.05). Dual-luciferase reporter gene assay confirmed that NDRG2 was a target gene of miR-301a, and si-NDRG2 could reverse the neuroprotective effect of miR-301a inhibitor in rats with cerebral I/R injury. Inhibiting miR-301a has a neuroprotective effect on rats with cerebral I/R injury to ameliorate cell apoptosis and inflammatory response through possibly targeting NDRG2.