Ensemble Approach to Combining Episode Prediction Models Using Sequential Circadian Rhythm Sensor Data from Mental Health Patients.

Managing mood disorders poses challenges in counseling and drug treatment, owing to limitations. Counseling is the most effective during hospital visits, and the side effects of drugs can be burdensome. Patient empowerment is crucial for understanding and managing these triggers. The daily monitoring of mental health and the utilization of episode prediction tools can enable self-management and provide doctors with insights into worsening lifestyle patterns. In this study, we test and validate whether the prediction of future depressive episodes in individuals with depression can be achieved by using lifelog sequence data collected from digital device sensors. Diverse models such as random forest, hidden Markov model, and recurrent neural network were used to analyze the time-series data and make predictions about the occurrence of depressive episodes in the near future. The models were then combined into a hybrid model. The prediction accuracy of the hybrid model was 0.78; especially in the prediction of rare episode events, the F1-score performance was approximately 1.88 times higher than that of the dummy model. We explored factors such as data sequence size, train-to-test data ratio, and class-labeling time slots that can affect the model performance to determine the combinations of parameters that optimize the model performance. Our findings are especially valuable because they are experimental results derived from large-scale participant data analyzed over a long period of time.

ResBiGAAT: Residual Bi-GRU with attention for protein-ligand binding affinity prediction.

Protein-ligand interaction plays a crucial role in drug discovery, facilitating efficient drug development and enabling drug repurposing. Several computational algorithms, such as Graph Neural Networks and Convolutional Neural Networks, have been proposed to predict the binding affinity using the three-dimensional structure of ligands and proteins. However, there are limitations due to the need for experimental characterization of the three-dimensional structure of protein sequences, which is still lacking for some proteins. Moreover, these models often suffer from unnecessary complexity, resulting in extraneous computations. This study presents ResBiGAAT, a novel deep learning model that combines a deep Residual Bidirectional Gated Recurrent Unit with two-sided self-attention mechanisms. ResBiGAAT leverages protein and ligand sequence-level features and their physicochemical properties to efficiently predict protein-ligand binding affinity. Through rigorous evaluation using 5-fold cross-validation, we demonstrate the performance of our proposed approach. The model exhibits competitive performance on an external dataset, highlighting its generalizability. Our publicly available web interface, located at resbigaat.streamlit.app, allows users to conveniently input protein and ligand sequences to estimate binding affinity.

Melittin derived peptide-drug conjugate, M-DM1, inhibits tumor progression and induces effector cell infiltration in melanoma by targeting M2 tumor-associated macrophages.

Background: Melanoma has the highest mortality rate among all the types of skin cancer. In melanoma, M2-like tumor-associated macrophages (TAMs) are associated with the invasiveness of tumor cells and a poor prognosis. Hence, the depletion or reduction of M2-TAMs is a therapeutic strategy for the inhibition of tumor progression. The aim of this study was to evaluate the therapeutic effects of M-DM1, which is a conjugation of melittin (M), as a carrier for M2-like TAMs, and mertansine (DM1), as a payload to induce apoptosis of TAMs, in a mouse model of melanoma. Methods: Melittin and DM1 were conjugated and examined for the characterization of M-DM1 by high-performance liquid chromatography and electrospray ionization mass spectrometry. Synthesized M-DM1 were examined for in vitro cytotoxic effects. For the in vivo study, we engrafted murine B16-F10 into right flank of C57BL/6 female mice and administered an array of treatments (PBS, M, DM1, or M-DM1 (20 nmol/kg)). Subsequently, the tumor growth and survival rates were analyzed, as well as examining the phenotypes of tumor-infiltrating leukocytes and expression profiles. Results: M-DM1 was found to specifically reduce M2-like TAMs in melanoma, which potentially leads to the suppression of tumor growth, migration, and invasion. In addition, we also found that M-DM1 improved the survival rates in a mouse model of melanoma compared to M or DM1 treatment alone. Flow cytometric analysis revealed that M-DM1 enhanced the infiltration of CD8+ cytotoxic T cells and natural killer cells (NK cells) in the tumor microenvironment. Conclusion: Taken together, our findings highlight that M-DM1 is a prospective agent with enhanced anti-tumor effects.

Human activity recognition for analyzing stress behavior based on Bi-LSTM.

BACKGROUND: Stress is one of the critical health factors that could be detected by Human Activity Recognition (HAR) which consists of physical and mental health. HAR can raise awareness of self-care and prevent critical situations. Recently, HAR used non-invasive wearable physiological sensors. Moreover, deep learning techniques are becoming a significant tool for analyzing health data. OBJECTIVE: In this paper, we propose a human lifelog monitoring model for stress behavior recognition based on deep learning, which analyses stress levels during activity. The proposed approach considers activity and physiological data for recognizing physical activity and stress levels. METHODS: To tackle these issues, we proposed a model that utilizes hand-crafted feature generation techniques compatible with a Bidirectional Long Short-Term Memory (Bi-LSTM) based method for physical activity and stress level recognition. We have used a dataset called WESAD, collected using wearable sensors for model evaluation. This dataset presented four levels of stress emotion, including baseline, amusement, stress, and meditation. RESULTS: The following results are from the hand-crafted feature approaches compatible with the bidirectional LSTM model. The proposed model achieves an accuracy of 95.6% and an F1-score of 96.6%. CONCLUSION: The proposed HAR model efficiently recognizes stress levels and contributes to maintaining physical and mental well-being.

Optimal Feature Selection-Based Dental Caries Prediction Model Using Machine Learning for Decision Support System.

The high frequency of dental caries is a major public health concern worldwide. The condition is common, particularly in developing countries. Because there are no evident early-stage signs, dental caries frequently goes untreated. Meanwhile, early detection and timely clinical intervention are required to slow disease development. Machine learning (ML) models can benefit clinicians in the early detection of dental cavities through efficient and cost-effective computer-aided diagnoses. This study proposed a more effective method for diagnosing dental caries by integrating the GINI and mRMR algorithms with the GBDT classifier. Because just a few clinical test features are required for the diagnosis, this strategy could save time and money when screening for dental caries. The proposed method was compared to recently proposed dental procedures. Among these classifiers, the suggested GBDT trained with a reduced feature set achieved the best classification performance, with accuracy, F1-score, precision, and recall values of 95%, 93%, 99%, and 88%, respectively. Furthermore, the experimental results suggest that feature selection improved the performance of the various classifiers. The suggested method yielded a good predictive model for dental caries diagnosis, which might be used in more imbalanced medical datasets to identify disease more effectively.

MMDCP: Multi-Modal Dental Caries Prediction for Decision Support System Using Deep Learning.

In recent years, healthcare has gained unprecedented attention from researchers in the field of Human health science and technology. Oral health, a subdomain of healthcare described as being very complex, is threatened by diseases like dental caries, gum disease, oral cancer, etc. The critical point is to propose an identification mechanism to prevent the population from being affected by these diseases. The large amount of online data allows scholars to perform tremendous research on health conditions, specifically oral health. Regardless of the high-performing dental consultation tools available in current healthcare, computer-based technology has shown the ability to complete some tasks in less time and cost less than when using similar healthcare tools to perform the same type of work. Machine learning has displayed a wide variety of advantages in oral healthcare, such as predicting dental caries in the population. Compared to the standard dental caries prediction previously proposed, this work emphasizes the importance of using multiple data sources, referred to as multi-modality, to extract more features and obtain accurate performances. The proposed prediction model constructed using multi-modal data demonstrated promising performances with an accuracy of 90%, F1-score of 89%, a recall of 90%, and a precision of 89%.

Anticancer effect of verteporfin on non-small cell lung cancer via downregulation of ANO1.

Anoctamin 1 (ANO1) is a calcium-activated chloride channel found in various cell types and is overexpressed in non-small cell lung cancer (NSCLC), a major cause of cancer-related mortality. With the rising interest in development of druggable compounds for NSCLC, there has been a corresponding rise in interest in ANO1, a novel drug target for NSCLC. However, as ANO1 inhibitors that have been discovered simultaneously exhibit both the functions of an inhibition of ANO1 channel as well as a reduction of ANO1 protein levels, it is unclear which of the two functions directly causes the anticancer effect. In this study, verteporfin, a chemical compound that reduces ANO1 protein levels was identified through high-throughput screening. Verteporfin did not inhibit ANO1-induced chloride secretion but reduced ANO1 protein levels in a dose-dependent manner with an IC50 value of ~300 nM. Moreover, verteporfin inhibited neither P2Y receptor-induced intracellular Ca2+ mobilization nor cystic fibrosis transmembrane conductance regulator (CFTR) channel activity, and molecular docking studies revealed that verteporfin bound to specific sites of ANO1 protein. Confirming that verteporfin reduces ANO1 protein levels, we then investigated the molecular mechanisms involved in its effect on NSCLC cells. Interestingly, verteporfin decreased ANO1 protein levels, the EGFR-STAT3 pathway as well as ANO1 mRNA expression. Verteporfin reduced the viability of ANO1-expressing cells (PC9, and gefitinib-resistant PC9) and induced apoptosis by increasing caspase-3 activity and PARP-1 cleavage. However, it did not affect hERG channel activity. These results show that the anticancer mechanism of verteporfin is caused via the down-regulation of ANO1.

Prediction of trehalose-metabolic pathway and comparative analysis of KEGG, MetaCyc, and RAST databases based on complete genome of Variovorax sp. PAMC28711.

BACKGROUND: Metabolism including anabolism and catabolism is a prerequisite phenomenon for all living organisms. Anabolism refers to the synthesis of the entire compound needed by a species. Catabolism refers to the breakdown of molecules to obtain energy. Many metabolic pathways are undisclosed and many organism-specific enzymes involved in metabolism are misplaced. When predicting a specific metabolic pathway of a microorganism, the first and foremost steps is to explore available online databases. Among many online databases, KEGG and MetaCyc pathway databases were used to deduce trehalose metabolic network for bacteria Variovorax sp. PAMC28711. Trehalose, a disaccharide, is used by the microorganism as an alternative carbon source. RESULTS: While using KEGG and MetaCyc databases, we found that the KEGG pathway database had one missing enzyme (maltooligosyl-trehalose synthase, EC 5.4.99.15). The MetaCyc pathway database also had some enzymes. However, when we used RAST to annotate the entire genome of Variovorax sp. PAMC28711, we found that all enzymes that were missing in KEGG and MetaCyc databases were involved in the trehalose metabolic pathway. CONCLUSIONS: Findings of this study shed light on bioinformatics tools and raise awareness among researchers about the importance of conducting detailed investigation before proceeding with any further work. While such comparison for databases such as KEGG and MetaCyc has been done before, it has never been done with a specific microbial pathway. Such studies are useful for future improvement of bioinformatics tools to reduce limitations.

GCRNN: graph convolutional recurrent neural network for compound-protein interaction prediction.

BACKGROUND: Compound-protein interaction prediction is necessary to investigate health regulatory functions and promotes drug discovery. Machine learning is becoming increasingly important in bioinformatics for applications such as analyzing protein-related data to achieve successful solutions. Modeling the properties and functions of proteins is important but challenging, especially when dealing with predictions of the sequence type. RESULT: We propose a method to model compounds and proteins for compound-protein interaction prediction. A graph neural network is used to represent the compounds, and a convolutional layer extended with a bidirectional recurrent neural network framework, Long Short-Term Memory, and Gate Recurrent unit is used for protein sequence vectorization. The convolutional layer captures regulatory protein functions, while the recurrent layer captures long-term dependencies between protein functions, thus improving the accuracy of interaction prediction with compounds. A database of 7000 sets of annotated compound protein interaction, containing 1000 base length proteins is taken into consideration for the implementation. The results indicate that the proposed model performs effectively and can yield satisfactory accuracy regarding compound protein interaction prediction. CONCLUSION: The performance of GCRNN is based on the classification accordiong to a binary class of interactions between proteins and compounds The architectural design of GCRNN model comes with the integration of the Bi-Recurrent layer on top of CNN to learn dependencies of motifs on protein sequences and improve the accuracy of the predictions.

Predicting Site Energy Usage Intensity Using Machine Learning Models.

Climate change is a shift in nature yet a devastating phenomenon, mainly caused by human activities, sometimes with the intent to generate usable energy required in humankind's daily life. Addressing this alarming issue requires an urge for energy consumption evaluation. Predicting energy consumption is essential for determining what factors affect a site's energy usage and in turn, making actionable suggestions to reduce wasteful energy consumption. Recently, a rising number of researchers have applied machine learning in various fields, such as wind turbine performance prediction, energy consumption prediction, thermal behavior analysis, and more. In this research study, using data publicly made available by the Women in Data Science (WiDS) Datathon 2022 (contains data on building characteristics and information collected by sensors), after appropriate data preparation, we experimented four main machine learning methods (random forest (RF), gradient boost decision tree (GBDT), support vector regressor (SVR), and decision tree for regression (DT)). The most performant model was selected using evaluation metrics: root mean square error (RMSE) and mean absolute error (MAE). The reported results proved the robustness of the proposed concept in capturing the insight and hidden patterns in the dataset, and effectively predicting the energy usage of buildings.

LLAD: Life-Log Anomaly Detection Based on Recurrent Neural Network LSTM.

Life-Log is a term used for the daily monitoring of health conditions and recognizing anomalies from data generated by sensor devices. The development of smart sensors enables collection of health data, which can be considered as a solution to risks associated with personal healthcare by raising awareness regarding health conditions and wellness. Therefore, Life-Log analysis methods are important for real-life monitoring and anomaly detection. This study proposes a method for the improvement and combination of previous methods and techniques in similar fields to detect anomalies in health log data generated by various sensors. Recurrent neural networks with long short-term memory units are used for analyzing the Life-Log data. The results indicate that the proposed model performs more effectively than conventional health data analysis methods, and the proposed approach can yield a satisfactory accuracy in anomaly detection.

Sex-specific metabolic interactions between liver and adipose tissue in MCD diet-induced non-alcoholic fatty liver disease.

Higher susceptibility to metabolic disease in male exemplifies the importance of sexual dimorphism in pathogenesis. We hypothesized that the higher incidence of non-alcoholic fatty liver disease in males involves sex-specific metabolic interactions between liver and adipose tissue. In the present study, we used a methionine-choline deficient (MCD) diet-induced fatty liver mouse model to investigate sex differences in the metabolic response of the liver and adipose tissue. After 2 weeks on an MCD-diet, fatty liver was induced in a sex-specific manner, affecting male mice more severely than females. The MCD-diet increased lipolytic enzymes in the gonadal white adipose tissue (gWAT) of male mice, whereas it increased expression of uncoupling protein 1 and other brown adipocyte markers in the gWAT of female mice. Moreover, gWAT from female mice demonstrated higher levels of oxygen consumption and mitochondrial content compared to gWAT from male mice. FGF21 expression was increased in liver tissue by the MCD diet, and the degree of upregulation was significantly higher in the livers of female mice. The endocrine effect of FGF21 was responsible, in part, for the sex-specific browning of gonadal white adipose tissue. Collectively, these data demonstrated that distinctively female-specific browning of white adipose tissue aids in protecting female mice against MCD diet-induced fatty liver disease.

Life data monitoring and analysis model for personalized healthcare.

As the focus of personal healthcare shifts from patient treatment to early detection and prevention, it is becoming increasingly important to manage personal wellness in our daily lives. Personal health monitoring of physical activities and status can be used to show users the distribution of their daily activities, making it easier for people to assess their health, adopt better lifestyles, and potentially decrease the occurrence of chronic diseases. In this paper, we propose a CA5W1HOnto-based life data monitoring model that provides basic monitored information from various devices and ensures preventive and proactive service for personalized healthcare. Additionally, we propose a life data analysis method to correlate the self-monitoring of activities with the status of the human body.

Dynamic access control model for privacy preserving personalized healthcare in cloud environment.

When sharing and storing healthcare data in a cloud environment, access control is a central issue for preserving data privacy as a patient's personal health data may be accessed without permission from many stakeholders. Specifically, dynamic authorization for the access of data is required because personal health data is stored in cloud storage via wearable devices. Therefore, we propose a dynamic access control model for preserving the privacy of personal healthcare data in a cloud environment. The proposed model considers context information for dynamic access. According to the proposed model, access control can be dynamically determined by changing the context information; this means that even for a subject with the same role in the cloud, access permission is defined differently depending on the context information and access condition. Furthermore, we experiment the ability of the proposed model to provide correct responses by representing a dynamic access decision with real-life personalized healthcare system scenarios.

Carotenogenesis in Haematococcus lacustris: role of protein tyrosine phosphatases.

In the present study, we examined the inhibitory effects of protein tyrosine phosphatase (PTPase) inhibitors, including sodium orthovanadate (SOV), ammonium molybdate (AM), and iodoacetamide (IA), on cell growth, accumulation of astaxanthin, and PTPase activity in the photosynthetic algae Haematococcus lacustris. PTPase activity was assayed spectrophotometrically and was found to be inhibited 60% to 90% after treatment with the inhibitors. SOV markedly abolished PTPase activity, significantly activating the accumulation of astaxanthin. These data suggest that the accumulation of astaxanthin in H. lacustris results from the concerted actions of several PTPases.

Selective control of the Prorocentrum minimum harmful algal blooms by a novel algal-lytic bacterium Pseudoalteromonas haloplanktis AFMB-008041.

In this study, we examined the algal-lytic activities and biological control mechanisms of Pseudoalteromonas haloplanktis AFMB-08041, which was isolated from surface seawater obtained at Masan Bay in Korea. In addition, we assessed whether AFMB-08041 could be used as a biocontrol agent to regulate harmful dinoflagellate Prorocentrum minimum. From these experiments, we found that the inoculation of AFMB-08041 at a final density of 2.5 x 10(4) cfu ml(-1) caused P. minimum cells to degrade (>90%) within 5 days. The algal cells were lysed through an indirect attack by the AFMB-08041 bacterial strain. Our results also suggest that the algal-lytic compounds produced by AFMB-08041 may have beta-glucosidase activity. However, P. haloplanktis AFMB-08041 was not able to suppress the growth of other alga such as Alexandrium tamarense, Akashiwo sanguinea, Cochlodinium polykrikoides, Gymnodinium catenatum, and Heterosigma akashiwo. Moreover, we observed that the growth of Prorocentrum dentatum, which has a very similar morphological structure to P. minimum, was also effectively suppressed by P. haloplanktis AFMB-08041. Therefore, the effect of AFMB-08041 on P. minimum degradation appears to be species specific. When testing in an indoor mesocosms, P. haloplanktis AFMB-08041 reduced the amount of viable P. minimum cells by 94.5% within 5 days after inoculation. The combined results of this study clearly demonstrate that this bacterium is capable of regulating the harmful algal blooms of P. minimum. In addition, these results will enable us to develop a new strategy for the anthropogenic control of harmful algal bloom-forming species in nature.

Statistical optimization of the growth factors for Chaetoceros neogracile using fractional factorial design and central composite design.

Statistical experimental designs; involving (i) a fractional factorial design (FFD) and (ii) a central composite design (CCD) were applied to optimize the culture medium constituents for production of a unique antifreeze protein by the Antartic microalgae Chaetoceros neogracile. The results of the FFD suggested that NaCl, KCl, MgCl2, and Na2SiO3 were significant variables that highly influenced the growth rate and biomass production. The optimum culture medium for the production of an antifreeze protein from C. neogracile was found to be Kalleampersandrsquor;s artificial seawater, pH of 7.0ampersandplusmn;0.5, consisting of 28.566 g/l of NaCl, 3.887 g/l of MgCl2, 1.787 g/l of MgSO4, 1.308 g/l of CaSO4, 0.832 g/l of K2SO4, 0.124 g/l of CaCO3, 0.103 g/l of KBr, 0.0288 g/l of SrSO4, and 0.0282 g/l of H3BO3. The antifreeze activity significantly increased after cells were treated with cold shock (at -5oC) for 14 h. To the best of our knowledge, this is the first report demonstrating an antifreeze-like protein of C. neogracile.

Inhibitory Effect of Algal Extracts on Mycelial Growth of the Tomato-Wilt Pathogen, Fusarium oxysporum f. sp. lycopersici.

The present study was undertaken to explore the inhibitory effect of cyanobacterial extracts of Nostoc commune FA-103 against the tomato-wilt pathogen, Fusarium oxysporum f. sp. lycopersici. In an optimal medium, cell growth, antifungal activity, and antifungal compound production could be increased 2.7-fold, 4.1-fold, and 13.4-fold, respectively. A crude algal extract had a similar effect as mancozeb at the recommended dose, both in laboratory and pot tests. In vitro and in vivo fungal growth, spore sporulation and fungal infection of wilt pathogen in tomato seeds were significantly inhibited by cyanobacterial extracts. Nostoc commune FA-103 extracts have potential for the suppression of Fusarium oxysporum f. sp. lycopersici.

Purification and characterization of extracellular beta-glucosidase from Sinorhizobium kostiense AFK-13 and its algal lytic effect on Anabaena flos-aquae.

A beta-glucosidase from the algal lytic bacterium Sinorhizobium kostiense AFK-13, grown in complex media containing cellobiose, was purified to homogeneity by successive ammonium sulfate precipitation, and anion-exchange and gel-filtration chromatographies. The enzyme was shown to be a monomeric protein with an apparent molecular mass of 52 kDa and isoelectric point of approximately 5.4. It was optimally active at pH 6.0 and 40'C and possessed a specific activity of 260.4 U/mg of protein against 4-nitrophenyl-beta-D-glucopyranoside (pNPG). A temperature-stability analysis demonstrated that the enzyme was unstable at 50 degrees C and above. The enzyme did not require divalent cations for activity, and its activity was significantly suppressed by Hg+2 and Ag+, whereas sodium dodecyl sulfate (SDS) and Triton X-100 moderately inhibited the enzyme to under 70% of its initial activity. In an algal lytic activity analysis, the growth of cyanobacteria, such as Anabaena flos-aquae, A. cylindrica, A. macrospora, Oscillatoria sancta, and Microcystis aeruginosa, was strongly inhibited by a treatment of 20 ppm/disc or 30 ppm/disc concentration of the enzyme.

Inhibition of Aflatoxin Production of Aspergillus flavus by Lactobacillus casei.

Lactobacillus casei KC-324 was tested for its ability to inhibit aflatoxin production and mycelial growth of Aspergillus flavus ATCC 15517 in liquid culture. Aflatoxin B1 biosynthesis and mycelial growth were inhibited in both simultaneous culture and individual antagonism assays,suggesting that the inhibitory activity was due to extracellular metabolites produced in cell-free supernatant fluids of the cultured broth of L. casei KC-324. In cell-free supernatant fluids of all media tested,deMan,Rogosa and Sharpe broth,potato dextrose broth,and Czapek-Dox broth + 1% yeast extract showed higher antiaflatoxigenic activity. In these case, fungal growths, however, was not affected as measured by mycelial dry weight. The antiaflatoxigenic metabolites from L. casei KC-324 were produced over wide range of temperatures between 25℃ and 37℃. However, these metabolites were not thermostable since the inhibitory activity of the supernatant was inactivated within 30 minutes at 100℃ and 121℃. The inhibitory activity was not influenced by changing pH of supernatant between 4 and 10. However,the antiaflatoxigenic activity was slightly reduced at pH 10.