Current Science

Utilization of Silkworm Pupae-Derived Biochar for Cr(VI) Removal

2026-07-23T10:57:10+00:00

Silkworm pupae, a prevalent agricultural residue in China, are rich in proteins and polysaccharides, offering significant potential for resourceful utilization. This study investigates the application of silkworm pupa-derived biochar for the removal of hexavalent chromium [Cr(VI)] from water. Biochar was produced through pyrolysis at 300 °C, 400 °C, and 500 °C at varying retention times. The silkworm pupa biochar produced at 300 °C exhibited effective Cr(VI) removal under acidic conditions (pH < 3), at a relatively low temperature of 15 °C, and with an adsorbent dosage below 5 g/L, reaching a maximum adsorption capacity of 1.476 mg/g. The adsorption process was consistent with the pseudo-second-order kinetic equation and the Freundlich isotherm adsorption model, which indicate that the adsorption of Cr(VI) by the biochar proceeded simultaneously with membrane diffusion and intraparticle diffusion. Some oxygen-containing functional groups and C=C groups were involved in the adsorption process. The study highlights the feasibility of using silkworm pupa biochar as a sustainable and economically viable adsorbent for wastewater remediation. These findings provide a potential approach for utilizing agricultural waste in environmental management, contributing to both waste treatment and pollution control.

An Integrated Drone-Borne Magnetics and Multi-Scale Satellite Remote Sensing to Delineate Structural Controls on Groundwater Accumulation in Hard-Rock Aquifers of Ghumnur Khurd surroundings, Komaram Bheem Asifabad Dist., Telangana, India

2026-07-28T06:04:40+00:00

A high-resolution drone-borne magnetic survey was integrated with multi-scale satellite remote sensing over approximately 10 Km² of tribal agricultural land in Komaram Bheem Asifabad district, northern Telangana, India, to identify subsurface dolerite dykes and assess their potential as natural barriers for passive groundwater storage. A total of 23,432 magnetic readings were acquired at 10Hz using a UAV- mounted proton precession magnetometer at a mean altitude of 581m amsl. Total Magnetic Intensity (TMI) data were processed to generate residual anomaly, Reduction to Pole (RTP), and Analytic Signal maps. Three distinct dolerite dykes were delineated: two NW–SE trending bodies (Dyke A: +949nT; Dyke 2: +160nT) and one NE–SW cross-cutting lateral barrier (Dyke 3: +198nT). Dip directions and geometries were determined using the North–South magnetic gradient asymmetry method and independently validated by satellite terrain slope, geomorphology, and lineament density maps. Both NW–SE dykes dip northward at 65⁰–80°, directly opposing the regional NNE → SSW groundwater flow, creating highly effective natural subsurface dams. Peters’ half-width depth estimates place the top of the dyke intrusions at 30–55m, with basement depths varying between 75m and 134m along three modelled depth profiles. Total passive groundwater storage potential is estimated at 8.3 million litres, which scales to 11.1 million litres with targeted surface check dam integration. Eight priority borewell locations and five surface check dam sites are proposed to enhance community water security. Published regional data indicate acceptable water quality for drinking and irrigation; however, mandatory fluoride testing is required at each borehole prior to commissioning due to localized geogenic risks. This study demonstrates that integrated drone magnetic workflows, cross-validated by multi-band satellite imagery and digital terrain layers; provide a rapid, non-invasive, and cost-effective paradigm for groundwater resource exploration in complex hard-rock aquifers.

A New Method Based on Feature Alignment for Multi-Center Machine Learning Survival Prediction in Nasopharyngeal Cancer Patients

2026-08-08T00:56:54+00:00

Purpose In the survival prediction of multi-center nasopharyngeal carcinoma patients, the inconsistency of data characteristics across different centers significantly affects the predictive performance of data-driven machine learning methods. This study explores how to improve the effectiveness and consistency of multi-center data through feature alignment and fusion techniques, thereby optimizing the performance of survival prediction models.

Methods We collected two sets of nasopharyngeal cancer data sets.After feature alignment, LASSO regression was used to screen the features in the aligned latent space to select the most discriminative features. Subsequently, we introduced 15 mainstream machine learning algorithms for survival prediction model training, selecting multidimensional metrics such as AUC, F1-score, Accuracy, and Recall for evaluation.Additionally, to validate the superiority of the method, we applied Pearson correlation and the Boruta algorithm to train the aforementioned 15 machine learning models separately on both datasets and compared the performance results. Combining the encoder weights of an Autoencoder, the SHAP importance is back-projected from the latent space to the original feature variables to analyze the factors influencing the survival of NPC patients.

Results All evaluation metrics (AUC, F1 score, precision) exceeding 0.97.This approach successfully mitigated multi-centre data heterogeneity while validating the robust generalisation capability and stability of these models within multi-centre data environments.SHAP interpretability analysis revealed that the feature combinations of gender + time, age + time, and primary site + age exhibited the broadest SHAP value distributions, representing the most influential feature combinations affecting survival risk prediction.

Conclusion The data heterogeneity issue associated with multi-center NPC survival prediction was effectively resolved by combining autoencoder-based feature alignment with adversarial training.

Circular RNA circASAP1 facilitates gastric cancer progression by sponging miR-874-3p and upregulating CPEB4

2026-08-09T08:44:20+00:00

Gastric cancer (GC) ranks as the fifth most prevalent malignancy globally and the fifth leading cause of cancer-related mortality. Although age-standardized incidence rates have declined gradually over recent decades, the absolute case burden continues to rise, driven largely by population aging and demographic growth. The clinical outlook for advanced GC remains poor, creating an urgent need to dissect the underlying molecular mechanisms. Circular RNAs (circRNAs) are established post-transcriptional regulators across multiple cancer types; however, the functional roles of most circRNAs in GC pathogenesis remain poorly characterized. In this study, we identified a significantly upregulated circRNA, hsa_circ_0008934 (designated circASAP1), in GC tissues and cell lines through microarray profiling and qRT-PCR validation. Functional assays demonstrated that circASAP1 knockdown markedly suppressed, while its overexpression promoted, GC cell proliferation, colony formation, DNA synthesis, migration, and invasion in vitro. Mechanistically, circASAP1 acts as a competitive endogenous RNA (ceRNA) by directly sponging miR-874-3p, thereby derepressing the oncogenic target CPEB4. Luciferase reporter and RNA fluorescence in situ hybridization (FISH) assays confirmed the direct interactions among circASAP1, miR-874-3p, and the CPEB4 3′ untranslated region. Rescue experiments validated that the pro-tumorigenic effects of circASAP1 are mediated, at least in part, through the miR-874-3p/CPEB4 axis. These findings establish the circASAP1/miR-874-3p/CPEB4 regulatory cascade as a driver of GC progression and identify circASAP1 as a candidate prognostic biomarker and therapeutic target.