Cognitive diagnostic assessment: A Q-matrix constraint-based neural network method.

Cognitive diagnosis is a crucial element of intelligent education that aims to assess the proficiency of specific skills or traits in students at a refined level and provide insights into their strengths and weaknesses for personalized learning. Researchers have developed numerous cognitive diagnostic models. However, previous studies indicate that diagnostic accuracy can be significantly influenced by the appropriateness of the model and the sample size. Thus, designing a general model that can adapt to different assumptions and sample sizes remains a considerable challenge. Artificial neural networks have been proposed as a promising approach in some studies. In this paper, we propose a cognitive diagnosis model of a neural network constrained by a Q-matrix and named QNN. Specifically, we employ the Q-matrix to determine the connections between neurons and the width and depth of the neural network. Moreover, to reduce the human effort in the training algorithm, we designed a self-organizing map-based cognitive diagnosis training framework called SOM-NN, which enables the QNN to be trained unsupervised. Extensive experimental results on simulated and real datasets demonstrate that our approaches are effective in both accuracy and interpretability. Notably, under unsupervised conditions, our approach has significant advantages on small sample datasets with high levels of guessing and slipping, especially on the pattern-wise agreement rates. This work bridges the gap between psychometrics and machine learning and provides a realistic and implementable reference solution for classroom instructional assessment and the cold start of personalized and adaptive assessment systems.

Effect of Pretreatment and Drying on the Nutritional and Functional Quality of Raisins Produced with Seedless Purple Grapes.

Raisins, known for their delicious taste and high nutritional value, are among the most widely consumed dried fruits globally. The natural waxy layer on the surface of grapes impedes water migration, making pretreatment necessary before drying. This study evaluated the effects of various pretreatment methods on the nutritional and functional quality of seedless purple raisins. By using non-pretreated dry seedless purple raisins as a control, the impact of physical and chemical pretreatment methods on the nutritional and functional qualities of seedless purple raisins was assessed through the analysis of nutrient content, phenolic compounds, and antioxidant activity. Our results demonstrate that physical pretreatment significantly increases the levels of vitamin C, fructose, glucose, total acid, total phenolics, total flavonoids, total anthocyanins, and antioxidant activity compared to chemical pretreatment and the control group. The correlation analysis revealed that phenolic substances were closely linked to antioxidant capacity. Additionally, phenolic compounds, including resveratrol, ferulic acid, chlorogenic acid, ethyl coumarate, and cinnamic acid, were more abundant following physical pretreatment. The OPLS-DA model effectively differentiated the three groups of processed samples, showing that different pretreatments significantly affect the nutritional and functional quality of seedless purple raisins. These findings suggest that physical pretreatment offers considerable potential for improving the drying quality of seedless purple raisins.

Intravenous immunoglobulin protects the integrity of the intestinal epithelial barrier and inhibits ferroptosis induced by radiation exposure by activating the mTOR pathway.

Radiation exposure often leads to serious health problems in humans. The intestinal epithelium is sensitive to radiation damage, and radiation causes destruction of the intestinal epithelial barrier, which leads to radiation enteritis (RE), the loss of fluids, and the translocation of intestinal bacteria and toxins; radiation can even threaten survival. In this study, we aimed to explore the influence of IVIg on the integrity of the intestinal epithelial barrier after RE. Using a RE mouse model, we investigated the protective effects of intravenous immunoglobulin (IVIg) on the epithelial junctions of RE mice and validated these findings with intestinal organoids cultured in vitro. In addition, transmission electron microscopy (TEM), western blotting (WB) and immunostaining were used to further investigate changes in intestinal epithelial ferroptosis and related signaling pathways. When RE occurs, the intestinal epithelial barrier is severely damaged. IVIg treatment significantly ameliorated this damage to epithelial tight junctions both in vivo and in vitro. Notably, IVIg alleviated RE by inhibiting intestinal epithelial ferroptosis in RE mice. Mechanistically, IVIg promoted activation of the mTOR pathway and inhibited ferroptosis in the intestinal epithelium of mice. Rapamycin, which is a potent inhibitor of the mTOR protein, significantly abolished the protective effect of IVIg against radiation-induced damage to intestinal epithelial tight junctions. Overall, IVIg can prevent RE-induced damage to the intestinal epithelial barrier and inhibit ferroptosis by activating the mTOR pathway; this study provides a new treatment strategy for patients with RE caused by radiotherapy or accidental nuclear exposure.

Crude Lipopeptides Produced by Bacillus amyloliquefaciens Could Control the Growth of Alternaria alternata and Production of Alternaria Toxins in Processing Tomato.

Alternaria spp. and its toxins are the main contaminants in processing tomato. Based on our earlier research, the current study looked into the anti-fungal capacity of crude lipopeptides from B. amyloliquefaciens XJ-BV2007 against A. alternata. We found that the crude lipopeptides significantly inhibited A. alternata growth and reduced tomato black spot disease incidence. SEM analysis found that the crude lipopeptides could change the morphology of mycelium and spores of A. alternata. Four main Alternaria toxins were detected using UPLC-MS/MS, and the findings demonstrated that the crude lipopeptides could lessen the accumulation of Alternaria toxins in vivo and in vitro. Meanwhile, under the stress of crude lipopeptides, the expression of critical biosynthetic genes responsible for TeA, AOH, and AME was substantially down-regulated. The inhibitory mechanism of the crude lipopeptides was demonstrated to be the disruption of the mycelial structure of A. alternata, as well as the integrity and permeability of the membrane of A. alternata sporocytes. Taken together, crude lipopeptides extracted from B. amyloliquefaciens XJ-BV2007 are an effective biological agent for controlling tomato black spot disease and Alternaria toxins contamination.

Time-Series Transcriptome of Cucumis melo Reveals Extensive Transcriptomic Differences with Different Maturity.

As the most important melon cultivar grown in the north-western provinces of China, Hami melon (Cucumis melo) produces large edible fruits that serve as an important dietary component in the world. In general, as a climacteric plant, melon harvested at 60% maturity results in a product with bad quality, while the highest-quality product can be guaranteed when harvesting at 90% maturity. In order to clarify the genetic basis of their distinct profiles of metabolite accumulation, we performed systematic transcriptome analyses between 60% and 90% maturity melons. A total of 36 samples were sequenced and over 1.7 billion reads were generated. Differentially expressed genes in 60% and 90% maturity melons were detected. Hundreds of these genes were functionally enriched in the sucrose and citric acid accumulation process of C. melo. We also detected a number of distinct splicing events between 60% and 90% maturity melons. Many genes associated with sucrose and citric acid accumulation displayed as differentially expressed or differentially spliced between different degrees of maturity of Hami melons, including CmCIN2, CmSPS2, CmBGAL3, and CmSPS2. These results demonstrate that the phenotype pattern differences between 60% and 90% maturity melons may be largely resulted from the significant transcriptome regulation.

Britannin inhibits cell proliferation, migration and glycolysis by downregulating KLF5 in lung cancer.

Lung cancer is a harmful type of malignancy and the leading cause of cancer-associated mortality. It is therefore imperative to develop novel drugs effective for treating this cancer. The Traditional Chinese Medicine compound Britannin has been previously reported to inhibit the development of certain cancers, such as pancreatic, breast and liver cancer. Moreover, Kruppel-like factor 5 (KLF5) has been identified an on oncogene in lung cancer. In the present study, the possible regulatory effects and underlying mechanism of Britannin in lung cancer were investigated. A549 and 16HBE cells were treated with different concentrations of Britannin. Subsequently, Cell counting kit-8, EdU staining and colony formation assays were used to detect the proliferative ability of these cells. Cell migration was detected by wound healing and Transwell assays, respectively. XF96 extracellular flux analyzer was used to analyze the extent of extracellular acidification and oxygen consumption rate in cells, whereas assay kits were used to detect glucose and lactic acid levels in the cell supernatant. The targeting effect between Britannin and the KLF5 protein was investigated using molecular docking technology. The protein expression levels of KLF5 in cells challenged with Britannin was detected by western blotting. Finally, overexpression of KLF5 in A549 cells was performed before cell proliferation, migration and the glycolysis rate were measured to explore the regulatory effects of Britannin. Britannin was found to inhibit the proliferation, migration and glycolysis of lung cancer cells, during which the protein expression levels of KLF5 were decreased. This suggests that Britannin regulated the expression of KLF5 in A549 cells. Overexpression of KLF5 reversed the inhibitory effects of Britannin on the proliferation, migration and glycolysis in lung cancer cells. In conclusion, these results suggest that Britannin can inhibit cell proliferation, migration and glycolysis by downregulating KLF5 expression in lung cancer cells.

Foodborne Pathogen and Microbial Community Differences in Fresh Processing Tomatoes in Xinjiang, China.

The microbes on fresh processing tomatoes correlate closely with diseases, preservation, and quality control. Investigation of the microbial communities on processing tomatoes from different production regions may help define microbial specificity, inform disease prevention methods, and improve quality. In this study, surface microbes on processing tomatoes from 10 samples in two primary production areas of southern and northern Xinjiang were investigated by sequencing fungal internal transcribed spacer and bacterial 16S rRNA hypervariable sequences. A total of 133 different fungal and bacterial taxonomies were obtained from processing tomatoes in the two regions, of which 63 genera were predominant. Bacterial and fungal communities differed significantly between southern and northern Xinjiang, and fungal diversity was higher in southern Xinjiang. Alternaria and Cladosporium on processing tomatoes in southern Xinjiang were associated with plant pathogenic risk. The plant pathogenic fungi of processing tomatoes in northern Xinjiang were more abundant in Alternaria and Fusarium. The abundance of Alternaria on processing tomatoes was higher in four regions of northern Xinjiang, indicating that there is a greater risk of plant pathogenicity in these areas. Processing tomatoes in northern and southern Xinjiang contained bacterial genera identified as gut microbes, such as Pantoea, Erwinia, Enterobacter, Enterococcus, and Serratia, indicating the potential risk of contamination of processing tomatoes with foodborne pathogens. This study highlighted the microbial specificity of processing tomatoes in two tomato production regions, providing a basis for further investigation and screening for foodborne pathogenic microorganisms.

Study on the Treatment of ITP Mice with IVIG Sourced from Distinct Sex-Special Plasma (DSP-IVIG).

Intravenous immunoglobulin (IVIG) is a first-line drug prepared from human plasma for the treatment of autoimmune diseases (AIDs), especially immune thrombocytopenia (ITP). Significant differences exist in protein types and expression levels between male and female plasma, and the prevalence of autoimmune diseases varies between sexes. The present study seeks to explore potential variations in IVIG sourced from distinct sex-specific plasma (DSP-IVIG), including IVIG sourced from female plasma (F-IVIG), IVIG sourced from male plasma (M-IVIG), and IVIG sourced from a blend of male and female plasma (Mix-IVIG). To address this question, we used an ITP mouse model and a monocyte-macrophage inflammation model treated with DSP IVIG. The analysis of proteomics in mice suggested that the pathogenesis and treatment of ITP may involve FcγRs mediated phagocytosis, apoptosis, Th17, cytokines, chemokines, and more. Key indicators, including the mouse spleen index, CD16+ macrophages, M1, M2, IL-6, IL-27, and IL-13, all indicated that the efficacy in improving ITP was highest for M-IVIG. Subsequent cell experiments revealed that M-IVIG exhibited a more potent ability to inhibit monocyte phagocytosis. It induced more necrotic M2 cells and fewer viable M2, resulting in weaker M2 phagocytosis. M-IVIG also demonstrated superiority in the downregulation of surface makers CD36, CD68, and CD16 on M1 macrophages, a weaker capacity to activate complement, and a stronger binding ability to FcγRs on the THP-1 surface. In summary, DSP-IVIG effectively mitigated inflammation in ITP mice and monocytes and macrophages. However, M-IVIG exhibited advantages in improving the spleen index, regulating the number and typing of M1 and M2 macrophages, and inhibiting macrophage-mediated inflammation compared to F-IVIG and Mix-IVIG.

EnsembleDL-ATG: Identifying autophagy proteins by integrating their sequence and evolutionary information using an ensemble deep learning framework.

Autophagy is a primary mechanism for maintaining cellular homeostasis. The synergistic actions of autophagy-related (ATG) proteins strictly regulate the whole autophagic process. Therefore, accurate identification of ATGs is a first and critical step to reveal the molecular mechanism underlying the regulation of autophagy. Current computational methods can predict ATGs from primary protein sequences, but owing to the limitations of algorithms, significant room for improvement still exists. In this research, we propose EnsembleDL-ATG, an ensemble deep learning framework that aggregates multiple deep learning models to predict ATGs from protein sequence and evolutionary information. We first evaluated the performance of individual networks for various feature descriptors to identify the most promising models. Then, we explored all possible combinations of independent models to select the most effective ensemble architecture. The final framework was built and maintained by an organization of four different deep learning models. Experimental results show that our proposed method achieves a prediction accuracy of 94.5 % and MCC of 0.890, which are nearly 4 % and 0.08 higher than ATGPred-FL, respectively. Overall, EnsembleDL-ATG is the first ATG machine learning predictor based on ensemble deep learning. The benchmark data and code utilized in this study can be accessed for free at https://github.com/jingry/autoBioSeqpy/tree/2.0/examples/EnsembleDL-ATG.

Sleep Deprivation and Heart Rate Variability in Healthy Volunteers: Effects of REM and SWS Sleep Deprivation.

Objective: Using PSG-guided acute selective REM/SWS sleep deprivation in volunteers, this study examined the effects of sleep deprivation on the cardiovascular and autonomic nervous systems, as well as the relationship between cardiac neuromodulation homeostasis and cardiovascular disease. Methods: An experiment was conducted using 30 healthy volunteers (male : female = 1 : 1, aged 26.33 ± 4.5 years) divided into groups for sleep deprivation of SWS and REM sleep, and then, each group was crossed over for normal sleep (2 days) and repeated sleep deprivation (1 day, 3 times). During the study period, PSG and ELECTRO ECG monitoring were conducted, and five-minute frequency domain parameters and blood pressure values were measured before and after sleep deprivation. Results: Changes in VLF, LFnu, LF/HF, HF, and HFnu after SWS sleep deprivation were statistically significant (P < 0.05), but not LF (P = 0.063). Changes in VLF, LF, HF, LF/HF, LFnu, and HFnu after REM sleep deprivation were not statistically significant (P > 0.05). Conclusions: An increase in sympathetic nerve activity results from sleep deprivation and sudden awakening from SWS sleep is associated with a greater risk of cardiovascular disease.

Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy.

Post-transcriptionally RNA modifications, also known as the epitranscriptome, play crucial roles in the regulation of gene expression during development. Recently, deep learning (DL) has been employed for RNA modification site prediction and has shown promising results. However, due to the lack of relevant studies, it is unclear which DL architecture is best suited for some pyrimidine modifications, such as 5-methyluridine (m5U). To fill this knowledge gap, we first performed a comparative evaluation of various commonly used DL models for epigenetic studies with the help of autoBioSeqpy. We identified optimal architectural variations for m5U site classification, optimizing the layer depth and neuron width. Second, we used this knowledge to develop Deepm5U, an improved convolutional-recurrent neural network that accurately predicts m5U sites from RNA sequences. We successfully applied Deepm5U to transcriptomewide m5U profiling data across different sequencing technologies and cell types. Third, we showed that the techniques for interpreting deep neural networks, including LayerUMAP and DeepSHAP, can provide important insights into the internal operation and behavior of models. Overall, we offered practical guidance for the development, benchmark, and analysis of deep learning models when designing new algorithms for RNA modifications.

Osimertinib combined with bevacizumab as the first-line treatment in non-small cell lung cancer patients with brain metastasis harboring epidermal growth factor receptor mutations.

BACKGROUND: The efficacy and safety of osimertinib combined with bevacizumab in non-small cell lung cancer (NSCLC) patients with brain metastasis harboring epidermal growth factor receptor (EGFR) mutations have not been fully studied. METHODS: Treatment-naïve NSCLC patients with brain metastasis harboring EGFR-activating mutations were treated with osimertinib 80 mg oral daily and bevacizumab 15 mg/kg intravenously on day 1, repeated every 21 days, until disease progression, intolerable toxicity, or death. The primary endpoint was the median progression-free survival (mPFS), and the secondary endpoints were the median overall survival (mOS), response rates, and toxicities. This study has been registered in ClinicalTrials.gov (NCT05104281) and is ongoing. RESULTS: A total of 52 Chinese patients were enrolled, of whom 17 harbored EGFR 19 del and 35 harbored EGFR L858R mutation. The objective response rate (ORR) was 75.0% and the disease control rate (DCR) was 96.2%; the mPFS was 17.0 months (95% CI: 11.46-22.54), while the mOS was not reached. The mPFS was 20.0 months (95% CI: 14.56-25.44) and was 17.0 months (95% CI: 13.28-20.72) for patients harboring EGFR 19 del and EGFR L858R mutation (p = 0.844), respectively. The intracranial ORR was 82.7%, and the intracranial mPFS was 22.0 months (95% CI: 2.92-41.08).The main adverse events were mild-to-moderate hand-foot syndrome, diarrhea, hypertension, and proteinuria. Three patients developed grade III proteinuria, while five patients developed grade III hypertension; they permanently discontinued bevacizumab treatment. CONCLUSIONS: Osimertinib combined with bevacizumab shows promising results in EGFR-mutated NSCLC patients with brain metastasis, and the side effects are tolerable.

Addressing the Conflict between Mobility and Stability in Oxide Thin-film Transistors.

Amorphous oxide semiconductor thin-film transistors (AOS TFTs) are ever-increasingly utilized in displays. However, to bring high mobility and excellent stability together is a daunting challenge. Here, the carrier transport/relaxation bilayer stacked AOS TFTs are investigated to solve the mobility-stability conflict. The charge transport layer (CTL) is made of amorphous In-rich InSnZnO, which favors big average effective coordination number for all cations and more edge-shared structures for better charge transport. Praseodymium-doped InSnZnO is used as the charge relaxation layer (CRL), which substantially shortens the photoelectron lifetime as revealed by femtosecond transient absorption spectroscopy. The CTL and CRL with the thickness suitable for industrial production respectively afford minute potential barrier fluctuation for charge transport and fast relaxation for photo-generated carriers, resulting in transistors with an ultrahigh mobility (75.5 cm2 V-1 s-1 ) and small negative-bias-illumination-stress/positive-bias-temperature-stress voltage shifts (-1.64/0.76 V). The design concept provides a promising route to address the mobility-stability conflict for high-end displays.

Implementation of the HACCP System for Apple Juice Concentrate Based on Patulin Prevention and Control.

Patulin (PAT) is a toxic secondary metabolite produced by Aspergillus sp. and Penicillium sp., which acts as a contaminant of most apples and their products. The internationally recognized HACCP system is selected as the theoretical basis to more effectively reduce the PAT in apple juice concentrate (AJC). Through field investigation of apple juice concentrate (AJC) production enterprises, we collected 117 samples from 13 steps of AJC production, including whole apple, apple pulp, and apple juice. PAT contents were analyzed via high-performance liquid chromatography (HPLC) and compared with samples from the different production processes. The result demonstrated that the PAT content was significantly (p < 0.05) influenced by five processes, receipt of raw apples, sorting of raw apples, adsorption step, pasteurization, and aseptic filling. These processes were determined as the CCPs. Monitoring systems for maintaining CCPs within acceptable limits were established, and corrective actions were proposed in case a CCP was surpassed. Based on the above-identified CCPs, critical limits, and control methods (corrective actions), a HACCP plan related to the production process of AJC was established. This study provided important guidance for juice manufacturers wishing to effectively control the PAT content in their products.

Effects of Bacillus amyloliquefaciens XJ-BV2007 on Growth of Alternaria alternata and Production of Tenuazonic Acid.

Large amounts of processing tomato are grown in Xinjiang, China. Tomato black spot disease, caused by Alternaria spp., and the produced alternaria toxins in tomato products are posing risks to human health. In this study, we isolated a rhizospheric bacterium, XJ-BV2007, from tomato (Solanum lycopersicum) fields, which we identified as Bacillus amyloliquefaciens. We found that this bacterium has a strong antagonistic effect against Alternaria alternata and reduces the accumulation of alternaria toxins in tomatoes. According to the antifungal activity of the bacteria-free filtrate, we revealed that B. amyloliquefaciens XJ-BV2007 suppresses A. alternata by the production of antifungal metabolites. Combining semi-preparative high-performance liquid chromatography, we employed UPLC-QTOF-MS analysis and the Oxford cup experiment to find that fengycin plays an important role in inhibiting A. alternata. This paper firstly reported that B. amyloliquefaciens efficiently controls tomato black spot disease and mycotoxins caused by A. alternata. B. amyloliquefaciens XJ-BV2007 may provide an alternative biocontrol strain for the prevention of tomato black spot disease.

Application of Metabolomics and Traditional Chinese Medicine for Type 2 Diabetes Mellitus Treatment.

Diabetes is a major global public health problem with high incidence and case fatality rates. Traditional Chinese medicine (TCM) is used to help manage Type 2 Diabetes Mellitus (T2DM) and has steadily gained international acceptance. Despite being generally accepted in daily practice, the TCM methods and hypotheses for understanding diseases lack applicability in the current scientific characterization systems. To date, there is no systematic evaluation system for TCM in preventing and treating T2DM. Metabonomics is a powerful tool to predict the level of metabolites in vivo, reveal the potential mechanism, and diagnose the physiological state of patients in time to guide the follow-up intervention of T2DM. Notably, metabolomics is also effective in promoting TCM modernization and advancement in personalized medicine. This review provides updated knowledge on applying metabolomics to TCM syndrome differentiation, diagnosis, biomarker discovery, and treatment of T2DM by TCM. Its application in diabetic complications is discussed. The combination of multi-omics and microbiome to fully elucidate the use of TCM to treat T2DM is further envisioned.

Proteomics profiles of blood glucose-related proteins involved in a Chinese longevity cohort.

BACKGROUND: High blood glucose level is one of the main characteristics of diabetes mellitus. Based on previous studies, it is speculated longevity families may have certain advantages in blood glucose regulation. However, limited information on these items has been reported. The purpose of this study was to profile differences of plasma proteomics between longevity subjects (with normal fructosamine (FUN) level) and non-longevity area participants (with exceeding standard FUN level). METHODS: In this study, a TMT-based proteomics analysis was used to profile differences of plasma proteomics between longevity subjects (with normal FUN level) and non-longevity area participants (with exceeding standard FUN level). Results were validated by Luminex detection. RESULTS: A total of 155 differentially expressed proteins (DEPs) were identified between these two groups. The DEPs related to blood glucose regulation were mainly involved in glycolysis/gluconeogenesis, pyruvate metabolism and propanoate metabolism, and most of the DEPs were contained in carbohydrate metabolism, PI3K-Akt pathway, glucagon signaling pathway and inflammatory response. Validation by Luminex detection confirmed that CD163 was down-regulated, and SPARC, PARK 7 and IGFBP-1 were up-regulated in longevity participants. CONCLUSIONS: This study not only highlighted carbohydrate metabolism, PI3K-Akt pathway, glucagon signaling pathway and inflammatory response may play important roles in blood glucose regulation, but also indicated that YWHAZ, YWHAB, YWHAG, YWHAE, CALM3, CRP, SAA2, PARK 7, IGFBP1 and VNN1 may serve as potential biomarkers for predicting abnormal blood glucose levels.

The Potential of Alternaria Toxins Production by A. alternata in Processing Tomatoes.

As a filamentous and spoilage fungus, Alternaria spp. can not only infect processing tomatoes, but also produce a variety of mycotoxins which harm the health of human beings. To explore the production of Alternaria toxins in processing tomatoes during growth and storage, four main Alternaria toxins and four conjugated toxins were detected by ultrahigh-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) and ultra-performance liquid chromatography-ion mobility quadrupole time-of-flight mass spectrometry (UPLC-IMS QToF MS) in processing tomatoes on different days after being inoculated with A. alternata. The results show that the content of Alternaria toxins in an in vivo assay is higher than that under field conditions. Tenuazonic acid (TeA) is the predominant toxin detected in the field (205.86~41,389.19 µg/kg) and in vivo (7.64~526,986.37 µg/kg) experiments, and the second-most abundant toxin is alternariol (AOH). In addition, a small quantity of conjugated toxins, AOH-9-glucoside (AOH-9-Glc) and alternariol monomethyl ether-3-glucoside (AME-3-Glc), were screened in the in vivo experiment. This is the first time the potential of Alternaria toxins produced in tomatoes during the harvest period has been studied in order to provide data for the prevention and control of Alternaria toxins.

Analysis of influencing factors of serum total protein and serum calcium content in plasma donors.

Background and objectives: The adverse effects of plasma donation on the body has lowered the odds of donation. The aim of this study was to investigate the prevalence of abnormal serum calcium and total serum protein related to plasma donation, identify the influencing factors, and come up with suggestions to make plasma donation safer. Methods: Donors from 10 plasmapheresis centers in five provinces of China participated in this study. Serum samples were collected before donation. Serum calcium was measured by arsenazo III colorimetry, and the biuret method was used for total serum protein assay. An automatic biochemical analyzer was used to conduct serum calcium and total serum protein tests. Results: The mean serum calcium was 2.3 ± 0.15 mmol/L and total serum protein was 67.75 ± 6.02 g/L. The proportions of plasma donors whose serum calcium and total serum protein were lower than normal were 20.55% (815/3,966) and 27.99% (1,111/3,969), respectively. There were significant differences in mean serum calcium and total serum protein of plasma donors with different plasma donation frequencies, gender, age, regions, and body mass index (BMI), (all p < 0.05). Logistic regression analysis revealed that donation frequencies, age, BMI and regions were significantly associated with a higher risk of low serum calcium level, and donation frequencies, gender, age and regions were significant determinants factors of odds of abnormal total serum protein. Conclusions: Donation frequencies, gender, age, regions, and BMI showed different effects on serum calcium and total serum protein. More attention should be paid to the age, donation frequency and region of plasma donors to reduce the probability of low serum calcium and low total serum protein.

The applications of deep learning algorithms on in silico druggable proteins identification.

INTRODUCTION: The top priority in drug development is to identify novel and effective drug targets. In vitro assays are frequently used for this purpose; however, traditional experimental approaches are insufficient for large-scale exploration of novel drug targets, as they are expensive, time-consuming and laborious. Therefore, computational methods have emerged in recent decades as an alternative to aid experimental drug discovery studies by developing sophisticated predictive models to estimate unknown drugs/compounds and their targets. The recent success of deep learning (DL) techniques in machine learning and artificial intelligence has further attracted a great deal of attention in the biomedicine field, including computational drug discovery. OBJECTIVES: This study focuses on the practical applications of deep learning algorithms for predicting druggable proteins and proposes a powerful predictor for fast and accurate identification of potential drug targets. METHODS: Using a gold-standard dataset, we explored several typical protein features and different deep learning algorithms and evaluated their performance in a comprehensive way. We provide an overview of the entire experimental process, including protein features and descriptors, neural network architectures, libraries and toolkits for deep learning modelling, performance evaluation metrics, model interpretation and visualization. RESULTS: Experimental results show that the hybrid model (architecture: CNN-RNN (BiLSTM) + DNN; feature: dictionary encoding + DC_TC_CTD) performed better than the other models on the benchmark dataset. This hybrid model was able to achieve 90.0% accuracy and 0.800 MCC on the test dataset and 84.8% and 0.703 on a nonredundant independent test dataset, which is comparable to those of existing methods. CONCLUSION: We developed the first deep learning-based classifier for fast and accurate identification of potential druggable proteins. We hope that this study will be helpful for future researchers who would like to use deep learning techniques to develop relevant predictive models.