Search

The incidence of intrahepatic cholangiocarcinoma (ICC) is steadily rising, and it is associated with a high mortality rate. Clinical samples were collected to detect the expression of HSPB8 and BAG3 in ICC tissues. ICC cells were cultured and transfected with plasmids that overexpressed or silenced specific genes to investigate the impact of gene expression alterations on cell function. qPCR and Western blot techniques were utilized to measure gene and protein expression levels. A wound healing assay was conducted to assess cell migration ability. The Transwell assay was used to assess cell invasion ability. Co-IP was used to verify the binding relationship between HSPB8 and BAG3. The effects of HSPB8 and BAG3 on lung metastasis of tumors in vivo were verified by constructing a metastatic tumor model. Through the above experiments, we discovered that the expressions of HSPB8 and BAG3 were up-regulated in ICC tissues and cells, and their expressions were positively correlated. The metastatic ability of ICC cells could be promoted or inhibited by upregulating or downregulating the expression of BAG3. Furthermore, the HSPB8-BAG3 chaperone complex resulted in the abnormal degradation of Filamin A by activating autophagy. Increased expression of Filamin A inhibits the migration and invasion of ICC cells. Overexpression of HSPB8 and BAG3 in vivo promoted the lung metastasis ability of ICC cells. The HSPB8-BAG3 chaperone complex promotes ICC cell migration and invasion by regulating CASA-mediated degradation of Filamin A, offering insights for enhancing ICC therapeutic strategies.

Named Entity Recognition (NER) is a fundamental task in natural language processing that aims to identify and categorize named entities within unstructured text. In recent years, with the development of deep learning techniques, pre-trained language models have been widely used in NER tasks. However, these models still face limitations in terms of their scalability and adaptability, especially when dealing with complex linguistic phenomena such as nested entities and long-range dependencies. To address these challenges, we propose the MacBERT-BiGRU-Self Attention-Global Pointer (MB-GAP) model, which integrates MacBERT for deep semantic understanding, BiGRU for rich contextual information, self-attention for focusing on relevant parts of the input, and a global pointer mechanism for precise entity boundary detection. By optimizing the number of attention heads and global pointer heads, our model achieves an effective balance between complexity and performance. Extensive experiments on benchmark datasets, including ResumeNER, CLUENER2020, and SCHOLAT-School, demonstrate significant improvements over baseline models.

To further enhance the industrial utilization value of Haematococcus lacustris, the plasma mutagenesis of Haematococcus lacustris was carried out by an atmospheric and room temperature plasma (ARTP) mutagenesis equipment. The optimum input power and mutagenesis time for plasma mutagenesis were determined with lethal rate of algal cells as the index. After mutagenesis, high-yield astaxanthin mutant algae strains were obtained through primary screening of solid plate culture and secondary screening of liquid culture. Then, the culture conditions of high yield algal plants at vegetative growth stage were optimized by single-factor and orthogonal experiment with algae cell density as the index, and the suitable high light conditions for astaxanthin accumulation during astaxanthin induction stage were selected. The genetic stability of the high yielding mutant algae strains was observed after multiple subcultures under the optimized culture conditions. The results showed that the optimum conditions for plasma mutagenesis of Haematococcus lacustris were 240 W for 150 s or 400 W for 120 s. 11 Mutant alga strains with fast growth and high astaxanthin yield were obtained through primary screening and rescreening, wherein the strain HP3 grew fastest and had the highest astaxanthin yield. After culture, its cell density and astaxanthin yield were increased by 25.5% and 61.6% respectively compared with the original strain. After two-stage optimization, the cell density and astaxanthin yield of HP3 increased by 14.3% and 19.3% respectively, reaching 7.2×105 cell/mL and 31.264 mg/L. HP3 showed good growth and stable heredity. Its cell density and astaxanthin yield were similar to those of primary culture. The results have practical significance for the breeding of industrial algal strains producing astaxanthin from Haematococcus lacustris.

The academic social networks (ASNs) play an important role in promoting scientific collaboration and innovation in academic society. Accompanying the tremendous growth of scholarly big data, finding suitable scholars on ASNs for collaboration has become more difficult. Different from friend recommendation in conventional social networks, scholar recommendation in ASNs usually involves different academic entities (e.g., scholars, scientific publications, and status updates) and various relationships (e.g., collaboration relationship between team members, citations, and co-authorships), which forms a complex heterogeneous academic network. Our goal is to recommend potential similar scholars for users in ASNs. In this article, we propose to design a graph embedding-based scholar recommendation system by leveraging academic auxiliary information. First, we construct enhanced ASNs by integrating two types of academic features extracted from scholars’ academic information with original network topology. Then, the refined feature representations of the scholars are obtained by a graph embedding framework, which helps the system measure the similarity between scholars based on their representation vectors. Finally, the system generates potential similar scholars for users in ASNs for the final recommendation. We evaluate the effectiveness of our model on five real-world datasets: SCHOLAT, Zhihu, APS, Yelp and Gowalla. The experimental results demonstrate that our model is effective and achieves promising improvements than the other competitive baselines.

The toxicity tests of nineteen commonly used pesticides were carried out to compare the sensitivity differences between predatory mite Neoseiulus cucumeris and its prey Tetranychus cinnabarinus by a “leaf spray method” in laboratory microcosms. For two avermectins, emamectin benzoate and abamectin, exhibited high bioactivity against T. cinnabarinusf with LR50 values of 0.04 and 0.05 g a.i./ha, respectively, but these two insecticides showed the opposite toxic effects to N. cucumeris. These two agents showed strong selectivity for the two test species with Selective Toxicity Rate (STR) values of 950 and 620, respectively. However, for five neonicotinoids, the LR50s of dinotefuran, thiamethoxam, imidacloprid, and acetamiprid were all greater than the recommended rates in the field except for clothianidin, and they showed no obvious toxicity difference to the two species with STR values ranging from 0.58 to 2.00. For two organophosphates, malathion is more toxic to N. cucumeris than T. cinnabarinus, however, dimethoate showed a higher toxic effect on T. cinnabarinus. In addition, the toxicity of four pyrethroids, bifenthrin, deltamethrin, cyhalothrin and gamma-cyhalothrin to N. cucumeris was higher than that of T. cinnabarinus, except for alpha-cypermethrin. For five acaricides, spirodiclofen, spirotetramat and pyridaben had no obvious selectivity to the two organisms, while diafenthiuron and chlorfenapyr were found to be highly toxic to T. cinnabarinus than N. cucumeris with STR values of 14.2 and 68.5, respectively. Thus, some pesticides above-mentioned like emamectin benzoate, abamectin, diafenthiuron and chlorfenapyr exhibited potential to be used in the management programs of T. cinnabarinus, especially in organically based production systems where there are fewer chemical control measures available, which need to combine with natural enemies to achieve the best control effect.

Academic procrastination has been reported affecting students' performance in computer-supported learning environments. Studies have shown that students who demonstrate higher procrastination tendencies achieve less than the students with lower procrastination tendencies. It is important for a teacher to be aware of the students' behaviors especially their procrastination trends. EDM techniques can be used to analyze data collected through computer-supported learning environments and to predict students' behaviors. In this paper, we present an algorithm called students' academic performance enhancement through homework late/non-submission detection (SAPE) for predicting students' academic performance. This algorithm is designed to predict students with learning difficulties through their homework submission behaviors. First, students are labeled as procrastinators or non-procrastinators using k-means clustering algorithm. Then, different classification methods are used to classify students using homework submission feature vectors. We use ten classification methods, i.e., ZeroR, OneR, ID3, J48, random forest, decision stump, JRip, PART, NBTree, and Prism. A detailed analysis is presented regarding performance of different classification methods for different number of classes. The analysis reveals that in general the prediction accuracy of all methods decreases with increase in the number of classes. However, different methods perform best or worst for different number of classes.