H2Opred: a robust and efficient hybrid deep learning model for predicting 2'-O-methylation sites in human RNA.

2'-O-methylation (2OM) is the most common post-transcriptional modification of RNA. It plays a crucial role in RNA splicing, RNA stability and innate immunity. Despite advances in high-throughput detection, the chemical stability of 2OM makes it difficult to detect and map in messenger RNA. Therefore, bioinformatics tools have been developed using machine learning (ML) algorithms to identify 2OM sites. These tools have made significant progress, but their performances remain unsatisfactory and need further improvement. In this study, we introduced H2Opred, a novel hybrid deep learning (HDL) model for accurately identifying 2OM sites in human RNA. Notably, this is the first application of HDL in developing four nucleotide-specific models [adenine (A2OM), cytosine (C2OM), guanine (G2OM) and uracil (U2OM)] as well as a generic model (N2OM). H2Opred incorporated both stacked 1D convolutional neural network (1D-CNN) blocks and stacked attention-based bidirectional gated recurrent unit (Bi-GRU-Att) blocks. 1D-CNN blocks learned effective feature representations from 14 conventional descriptors, while Bi-GRU-Att blocks learned feature representations from five natural language processing-based embeddings extracted from RNA sequences. H2Opred integrated these feature representations to make the final prediction. Rigorous cross-validation analysis demonstrated that H2Opred consistently outperforms conventional ML-based single-feature models on five different datasets. Moreover, the generic model of H2Opred demonstrated a remarkable performance on both training and testing datasets, significantly outperforming the existing predictor and other four nucleotide-specific H2Opred models. To enhance accessibility and usability, we have deployed a user-friendly web server for H2Opred, accessible at https://balalab-skku.org/H2Opred/. This platform will serve as an invaluable tool for accurately predicting 2OM sites within human RNA, thereby facilitating broader applications in relevant research endeavors.

Advancing the accuracy of SARS-CoV-2 phosphorylation site detection via meta-learning approach.

The worldwide appearance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has generated significant concern and posed a considerable challenge to global health. Phosphorylation is a common post-translational modification that affects many vital cellular functions and is closely associated with SARS-CoV-2 infection. Precise identification of phosphorylation sites could provide more in-depth insight into the processes underlying SARS-CoV-2 infection and help alleviate the continuing COVID-19 crisis. Currently, available computational tools for predicting these sites lack accuracy and effectiveness. In this study, we designed an innovative meta-learning model, Meta-Learning for Serine/Threonine Phosphorylation (MeL-STPhos), to precisely identify protein phosphorylation sites. We initially performed a comprehensive assessment of 29 unique sequence-derived features, establishing prediction models for each using 14 renowned machine learning methods, ranging from traditional classifiers to advanced deep learning algorithms. We then selected the most effective model for each feature by integrating the predicted values. Rigorous feature selection strategies were employed to identify the optimal base models and classifier(s) for each cell-specific dataset. To the best of our knowledge, this is the first study to report two cell-specific models and a generic model for phosphorylation site prediction by utilizing an extensive range of sequence-derived features and machine learning algorithms. Extensive cross-validation and independent testing revealed that MeL-STPhos surpasses existing state-of-the-art tools for phosphorylation site prediction. We also developed a publicly accessible platform at https://balalab-skku.org/MeL-STPhos. We believe that MeL-STPhos will serve as a valuable tool for accelerating the discovery of serine/threonine phosphorylation sites and elucidating their role in post-translational regulation.

High quality factor, monodisperse micron-sized random lasers based on porous PLGA spheres.

Miniature random lasers with high quality factor are crucial for applications in barcoding, bioimaging, and on-chip technologies. However, achieving monodisperse and size-tunable biocompatible random lasers has been a significant challenge. In this study, we employed poly(lactic-co-glycolic) acid (PLGA), a biocompatible material approved for medical use, as the base material for random lasers. By integrating a dye-doped PLGA solution with a microfluidic system, we successfully fabricated monodisperse and miniature dye-doped PLGA spheres with tunable sizes ranging from 25 to 52 µm. Upon optical pulse excitation, these spheres exhibited strong random lasing emission at 610-640 nm with a threshold of approximately 22 µJ·mm-2. The lasing modes demonstrated a spectral linewidth of 0.2 nm, corresponding to a quality factor of 3100. Fourier transform analysis of the lasing emission revealed fundamental cavity lengths, providing insights into the properties of the random lasers.

Predicting drought stress under climate change in the Southern Central Highlands of Vietnam.

In the Southern Central Highlands of Vietnam, droughts occur more frequently, causing significant damage and impacting the region's socio-economic development. During the dry season, rivers, streams, and reservoirs often face limited water availability, exacerbated in recent years by increasing drought severity. Recognizing the escalating severity of droughts, the study offers a novel contribution by conducting a comprehensive analysis of surface water resource distribution in Lam Dong province, focusing on assessing water demand for agricultural production, a crucial factor in ensuring sustainable crop growth. Two scenarios, Current-2020 (SC1) and Climate Change-2025 (SC2), are simulated, with SC2 based on climate change and sea level rise scenarios provided by the Ministry of Natural Resources and Environment (MONRE). These scenarios are integrated into the MIKE-NAM and MIKE-HYDRO basin models, allowing for a thorough assessment of the water balance of Lam Dong province. Furthermore, the study utilizes the Keetch-Byram Drought Index (KBDI) to measure drought severity, revealing prevalent dry and moderately droughty conditions in highland districts with rainfall frequency ranging from 50 to 85%. Severe drought conditions occur with a rainfall frequency of 95%, indicating an increased frequency and geographic scope of severe droughts. Additionally, the study highlights that under abnormally dry conditions, water demand for the winter-spring crop is consistently met at 100%, decreasing to 85%, 80%, and less than 75% for moderate, severe, and extreme droughts, respectively. These findings offer insights into future drought conditions in the Lam Dong province and their potential impact on irrigation capacity, crucial for adaptation strategies.

Fruit-CoV: An efficient vision-based framework for speedy detection and diagnosis of SARS-CoV-2 infections through recorded cough sounds.

COVID-19 is an infectious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This deadly virus has spread worldwide, leading to a global pandemic since March 2020. A recent variant of SARS-CoV-2 named Delta is intractably contagious and responsible for more than four million deaths globally. Therefore, developing an efficient self-testing service for SARS-CoV-2 at home is vital. In this study, a two-stage vision-based framework, namely Fruit-CoV, is introduced for detecting SARS-CoV-2 infections through recorded cough sounds. Specifically, audio signals are converted into Log-Mel spectrograms, and the EfficientNet-V2 network is used to extract their visual features in the first stage. In the second stage, 14 convolutional layers extracted from the large-scale Pretrained Audio Neural Networks for audio pattern recognition (PANNs) and the Wavegram-Log-Mel-CNN are employed to aggregate feature representations of the Log-Mel spectrograms and the waveform. Finally, the combined features are used to train a binary classifier. In this study, a dataset provided by the AICovidVN 115M Challenge is employed for evaluation. It includes 7,371 recorded cough sounds collected throughout Vietnam, India, and Switzerland. Experimental results indicate that the proposed model achieves an Area Under the Receiver Operating Characteristic Curve (AUC) score of 92.8% and ranks first on the final leaderboard of the AICovidVN 115M Challenge. Our code is publicly available.

Improving ligand-ranking of AutoDock Vina by changing the empirical parameters.

AutoDock Vina (Vina) achieved a very high docking-success rate, p^ , but give a rather low correlation coefficient, R , for binding affinity with respect to experiments. This low correlation can be an obstacle for ranking of ligand-binding affinity, which is the main objective of docking simulations. In this context, we evaluated the dependence of Vina R coefficient upon its empirical parameters. R is affected more by changing the gauss2 and rotation than other terms. The docking-success rate p^ is sensitive to the alterations of the gauss1, gauss2, repulsion, and hydrogen bond parameters. Based on our benchmarks, the parameter set1 has been suggested to be the most optimal. The testing study over 800 complexes indicated that the modified Vina provided higher correlation with experiment Rset1=0.556±0.025 compared with RDefault=0.493±0.028 obtained by the original Vina and RVina1.2=0.503±0.029 by Vina version 1.2. Besides, the modified Vina can be also applied more widely, giving R≥0.500 for 32/48 targets, compared with the default package, giving R≥0.500 for 31/48 targets. In addition, validation calculations for 1036 complexes obtained from version 2019 of PDBbind refined structures showed that the set1 of parameters gave higher correlation coefficient ( Rset1=0.617±0.017 ) than the default package ( RDefault=0.543±0.020 ) and Vina version 1.2 ( RVina1.2=0.540±0.020 ). The version of Vina with set1 of parameters can be downloaded at https://github.com/sontungngo/mvina. The outcomes would enhance the ranking of ligand-binding affinity using Autodock Vina.

Zoning the suitability of the western Mekong Delta for paddy rice cultivation and aquaculture under current and future environmental conditions.

Ca Mau and Kien Giang, the two provinces of the Mekong Delta bordering the Gulf of Thailand, are facing major environmental challenges affecting the agriculture and aquaculture sectors upon which many livelihoods in this region depend on. This study maps the suitability of these two provinces for paddy rice cultivation and shrimp farming according to soil characteristics and current and future environmental conditions for variables found to significantly influence the yield of those two sectors, i.e., the level of saltwater intrusion, water availability for rainfed agriculture, and the length of the growing period. Future environmental conditions were simulated using the MIKE 11 hydrodynamic model forced by four hydrodynamic scenarios, each one representing different extents of saltwater intrusion during both the dry and rainy seasons, while also considering the availability of water resources for rainfed agriculture. The suitability zoning was performed using a GIS-based analytic hierarchy process (AHP) approach, resulting in the categorisation of the land according to four suitability levels for each sector. The analysis reveals that paddy rice cultivation will become more suitable to Kien Giang province while shrimp farming will be more suitable to Ca Mau province if the simulated future environmental conditions materialise. A suitability analysis is essential for optimal utilisation of the land. The approach presented in this study will inform the regional economic development master plan and provide guidance to other delta regions experiencing severe environmental changes and wishing to consider potential future climatic and sea level changes, and their associated impacts, in their land use planning.

Monodisperse and size-tunable high-quality factor microsphere biolasers.

Microsphere biolasers have attracted a great deal of interest due to their potential for biosensing and cell tracking. Here we demonstrate a novel, to the best of our knowledge, microfluidic-based fabrication of nearly monodisperse dye-doped protein microsphere biolasers with a tunable size from 150 to 50 µm. In particular, for an 85 µm-bead, about 70% of the fabricated microspheres have the same size of 85 µm. Under optical pumping, the fabricated microspheres emit whispering gallery mode lasing emission with a lasing threshold of ${{7}}\;\unicode{x00B5} {\rm{J}}\;{{\rm{mm}}^{- 2}}$ and quality ($\!Q$) factor up to 3000. Interestingly, microspheres with the same size exhibit a similar lasing threshold and spectrum. The result indicates a high reproducibility of microsphere biolasers by the microfluidic-based fabrication technique. This Letter provides an effective method for mass production of high-$Q$ factor microsphere biolasers which is a significant step toward real biosensing and medical applications.

Environmental consequences of population, affluence and technological progress for European countries: A Malthusian view.

This paper examines the short-run and long-run effects of economic, sociological and energy factors on environmental degradation in 28 European countries. In so doing, we employ Panel Vector Autoregressive (PVAR) and Fully Modified OLS (FMOLS) approaches on data from 1990 to 2014 in a STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) framework. Key empirical results indicate that these factors may contribute to environmental improvement in the short run; however, there are adverse implications in the long-run. Specifically, economic factors including economic growth, trade openness and foreign direct investment cause environmental degradation in the under-analysis economies. The sociological factors as measured by the population growth and the level of urbanization also show a negative impact on the environmental degradation in the short-run but in the long run, both population size and urbanization increase environmental degradation. These findings are in line with the concerns raised by Thomas Robert Malthus in his Essay on the Principle of Population. With regards to the energy factors, it indicates that the renewable energies help the European environment by reducing the level of carbon dioxide emissions whereas the higher energy intensity is an ecological threat. Our results remain robust in the EKC framework.

Managing environmental challenges: Training as a solution to improve employee green performance.

Although environmental training program has become a vital solution to minimize environmental challenges, the existing literature has paid little attention to (1) the connection between environmental training and employee in-role green performance (EIGP), (2) the mediating role of employee environmental commitment in this environmental training-EIGP connection, and (3) the cultural perspective (Western and Local) influences on interrelationships between environmental training, EIGP, and employee environmental commitment. Drawing on the social exchange theory and cultural perspective, a quantitative study based on survey data collected from 301 respondents at hotels in Vietnam was employed to fill these research gaps. Findings reveal that environmental training program is as an important tool to drive EIGP directly, and this environmental training-EIGP relationship is significantly mediated by employee environmental commitment. Interestingly, the study indicates support for our prediction that the mediating role of employee environmental commitment on the environmental training-EIGP link is stronger at hotels managed by Western hospitality companies. However, unexpectedly, cultural influence does not moderate the effect of environmental training on EIGP. Other than theoretical contributions, our study carries important practical implications that can help organizations reduce their carbon footprint. Limitations and further research directions have also been discussed.

Protein-based microsphere biolasers fabricated by dehydration.

Biolasers made of biological materials have attracted considerable research attention due to their biocompatibility and biodegradability, and have the potential for biosensing and biointegration. However, the current fabrication methods of biolasers suffer from several limitations, such as complicated processing, time-consuming and environmentally unfriendly nature. In this study, a novel approach with green processes for fabricating solid-state microsphere biolasers has been demonstrated. By dehydration via a modified Microglassification™ technology, dye-doped bovine serum albumin (BSA) droplets could be quickly (less than 10 minutes) and easily changed into solid microspheres with diameters ranging from 10 µm to 150 µm. The size of the microspheres could be effectively controlled by changing either the concentration of the BSA solution or the diameter of the initial droplets. The fabricated microspheres could act as efficient microlasers under an optical pulse excitation. A lasing threshold of 7.8 µJ mm-2 and a quality (Q) factor of about 1700 to 3100 were obtained. The size dependence of lasing characteristics was investigated, and the results showed a good agreement with whispering gallery mode (WGM) theory. Our findings contribute an effective technique for the fabrication of high-Q factor microlasers that may be potential for applications in biological and chemical sensors.

Physical activity promotion for patients transitioning to dialysis using the "Exercise is Medicine" framework: a multi-center randomized pragmatic trial (EIM-CKD trial) protocol.

BACKGROUND: Patients on dialysis are physically inactive, with most reporting activity levels below the fifth percentile of healthy age-matched groups. Several small studies have reported efficacy of diverse exercise interventions among persons with CKD and those on dialysis. However, no single intervention has been widely adopted in real-world practice, despite a clear need in this vulnerable population with high rates of mortality, frailty, and skilled nursing hospitalizations. METHODS/DESIGN: We describe a pragmatic clinical trial for an exercise intervention among patients transitioning to dialysis. We will use an existing framework - Exercise is Medicine (EIM) - developed by the American College of Sports Medicine. After undertaking formative qualitative research to tailor the EIM framework to the advanced CKD population (eGFR < 30 ml/min/1.73m2), we will randomize 96 patients from two regions-Atlanta and Bay Area-in two intervention arms with incremental levels of clinical-community integration: physical activity assessment during Nephrology clinical visit, brief counseling at pre-dialysis education, and physical activity wearable (group 1) versus group 1 intervention components plus a referral to a free, EIM practitioner-led group exercise program over 16 weeks (group 2; 8 week core intervention; 8-week follow up). We will assess efficacy by comparing between group differences in minutes/week of objectively measured moderate intensity physical activity. To evaluate implementation, we will use questionnaires for assessing barriers to referral, participation and retention along the path of the intervention. Further we will have a plan for dissemination of the intervention by partnering with relevant stakeholders. DISCUSSION: The overall goal is to inform the development of a practical, cost-conscious intervention "package" that addresses barriers and challenges to physical activity commonly faced by patients with advanced CKD and can be disseminated amongst interested practices. TRIAL REGISTRATION: ClinicalTrials.gov identifier (Dated:10/17/2017): NCT03311763 .

Melanin-embedded materials effectively remove hexavalent chromium (CrVI) from aqueous solution.

BACKGROUND: Currently, it is recognized that water polluted with toxic heavy metal ions may cause serious effects on human health. Therefore, the development of new materials for effective removal of heavy metal ions from water is still a widely important area. Melanin is being considered as a potential material for removal of heavy metal from water. METHODS: In this study, we synthesized two melanin-embedded beads from two different melanin powder sources and named IMB (Isolated Melanin Bead originated from squid ink sac) and CMB (Commercial Melanin Bead originated from sesame seeds). These beads were of globular shape and 2-3 mm in diameter. We investigated and compared the sorption abilities of these two bead materials toward hexavalent-chromium (CrVI) in water. The isotherm sorption curves were established using Langmuir and Freundlich models in the optimized conditions of pH, sorption time, solid/liquid ratio, and initial concentration of CrVI. The FITR analysis was also carried out to show the differences in surface properties of these two beads. RESULTS: The optimized conditions for isotherm sorption of CrVI on IMB/CMB were set at pH values of 2/2, sorption times of 90/300 min, and solid-liquid ratios of 10/20 mg/mL. The maximum sorption capacities calculated based on the Langmuir model were 19.60 and 6.24 for IMB and CMB, respectively. However, the adsorption kinetic of CrVI on the beads fitted the Freundlich model with R2 values of 0.992 for IMB and 0.989 for CMB. The deduced Freundlich constant, 1/n, in the range of 0.2-0.8 indicated that these beads are good adsorption materials. In addition, structure analysis data revealed great differences in physical and chemical properties between IMB and CMB. Interestingly, FTIR analysis results showed strong signals of -OH (3295.35 cm- 1) and -C=O (1608.63 cm- 1) groups harboring on the IMB but not CMB. Moreover, loading of CrVI on the IMB caused a shift of broad peaks from 3295.35 cm- 1 and 1608.63 cm- 1 to 3354.21 cm- 1 and 1597.06 cm- 1, respectively, due to -OH and -C=O stretching. CONCLUSIONS: Taken together, our study suggests that IMB has great potential as a bead material for the elimination of CrVI from aqueous solutions and may be highly useful for water treatment applications.

Coupling High Throughput Microfluidics and Small-Angle X-ray Scattering to Study Protein Crystallization from Solution.

In this work, we propose the combination of small-angle X-ray scattering (SAXS) and high throughput, droplet based microfluidics as a powerful tool to investigate macromolecular interactions, directly related to protein solubility. For this purpose, a robust and low cost microfluidic platform was fabricated for achieving the mixing of proteins, crystallization reagents, and buffer in nanoliter volumes and the subsequent generation of nanodroplets by means of a two phase flow. The protein samples are compartmentalized inside droplets, each one acting as an isolated microreactor. Hence their physicochemical conditions (concentration, pH, etc.) can be finely tuned without cross-contamination, allowing the screening of a huge number of saturation conditions with a small amount of biological material. The droplet flow is synchronized with synchrotron radiation SAXS measurements to probe protein interactions while minimizing radiation damage. To this end, the experimental setup was tested with rasburicase (known to be very sensitive to denaturation), proving the structural stability of the protein in the droplets and the absence of radiation damage. Subsequently weak interaction variations as a function of protein saturation was studied for the model protein lysozime. The second virial coefficients (A2) were determined from the X-ray structure factors extrapolated to the origin. A2 obtained values were found to be in good agreement with data previously reported in literature but using only a few milligrams of protein. The experimental results presented here highlight the interest and convenience of using this methodology as a promising and potential candidate for studying protein interactions for the construction of phase diagrams.

Interspecies comparison of peptide substrate reporter metabolism using compartment-based modeling.

Peptide substrate reporters are fluorescently labeled peptides that can be acted upon by one or more enzymes of interest. Peptide substrates are readily synthesized and more easily separated than full-length protein substrates; however, they are often more rapidly degraded by peptidases. As a result, peptide reporters must be made resistant to proteolysis in order to study enzymes in intact cells and lysates. This is typically achieved by optimizing the reporter sequence in a single cell type or model organism, but studies of reporter stability in a variety of organisms are needed to establish the robustness and broader utility of these molecular tools. We measured peptidase activity toward a peptide substrate reporter for protein kinase B (Akt) in E. coli, D. discoideum, and S. cerevisiae using capillary electrophoresis with laser-induced fluorescence (CE-LIF). Using compartment-based modeling, we determined individual rate constants for all potential peptidase reactions and explored how these rate constants differed between species. We found the reporter to be stable in D. discoideum (t 1/2 = 82-103 min) and S. cerevisiae (t 1/2 = 279-314 min), but less stable in E. coli (t 1/2 = 21-44 min). These data suggest that the reporter is sufficiently stable to be used for kinase assays in eukaryotic cell types while also demonstrating the potential utility of compartment-based models in peptide substrate reporter design. Graphical abstract Cell lysates from several evolutionarily divergent species were incubated with a peptide substrate reporter, and compartment-based modeling was used to determine key steps in the metabolism of the reporter in each cell type.

Impact of peer support on virologic failure in HIV-infected patients on antiretroviral therapy - a cluster randomized controlled trial in Vietnam.

BACKGROUND: The effect of peer support on virologic and immunologic treatment outcomes among HIVinfected patients receiving antiretroviral therapy (ART) was assessed in a cluster randomized controlled trial in Vietnam. METHODS: Seventy-one clusters (communes) were randomized in intervention or control, and a total of 640 patients initiating ART were enrolled. The intervention group received peer support with weekly home-visits. Both groups received first-line ART regimens according to the National Treatment Guidelines. Viral load (VL) (ExaVir™ Load) and CD4 counts were analyzed every 6 months. The primary endpoint was virologic failure (VL >1000 copies/ml). Patients were followed up for 24 months. Intention-to-treat analysis was used. Cluster longitudinal and survival analyses were used to study time to virologic failure and CD4 trends. RESULTS: Of 640 patients, 71% were males, mean age 32 years, 83% started with stavudine/lamivudine/nevirapine regimen. After a mean of 20.8 months, 78% completed the study, and the median CD4 increase was 286 cells/µl. Cumulative virologic failure risk was 7.2%. There was no significant difference between intervention and control groups in risk for and time to virologic failure and in CD4 trends. Risk factors for virologic failure were ART-non-naïve status [aHR 6.9;(95% CI 3.2-14.6); p < 0.01]; baseline VL ≥100,000 copies/ml [aHR 2.3;(95% CI 1.2-4.3); p < 0.05] and incomplete adherence (self-reported missing more than one dose during 24 months) [aHR 3.1;(95% CI 1.1-8.9); p < 0.05]. Risk factors associated with slower increase of CD4 counts were: baseline VL ≥100,000 copies/ml [adj.sq.Coeff (95% CI): -0.9 (-1.5;-0.3); p < 0.01] and baseline CD4 count <100 cells/µl [adj.sq.Coeff (95% CI): -5.7 (-6.3;-5.4); p < 0.01]. Having an HIV-infected family member was also significantly associated with gain in CD4 counts [adj.sq.Coeff (95% CI): 1.3 (0.8;1.9); p < 0.01]. CONCLUSION: There was a low virologic failure risk during the first 2 years of ART follow-up in a rural low-income setting in Vietnam. Peer support did not show any impact on virologic and immunologic outcomes after 2 years of follow up. TRIAL REGISTRATION: NCT01433601 .

Tim-3 directly enhances CD8 T cell responses to acute Listeria monocytogenes infection.

T cell Ig and mucin domain (Tim) 3 is a surface molecule expressed throughout the immune system that can mediate both stimulatory and inhibitory effects. Previous studies have provided evidence that Tim-3 functions to enforce CD8 T cell exhaustion, a dysfunctional state associated with chronic stimulation. In contrast, the role of Tim-3 in the regulation of CD8 T cell responses to acute and transient stimulation remains undefined. To address this knowledge gap, we examined how Tim-3 affects CD8 T cell responses to acute Listeria monocytogenes infection. Analysis of wild-type (WT) mice infected with L. monocytogenes revealed that Tim-3 was transiently expressed by activated CD8 T cells and was associated primarily with acquisition of an effector phenotype. Comparison of responses to L. monocytogenes by WT and Tim-3 knockout (KO) mice showed that the absence of Tim-3 significantly reduced the magnitudes of both primary and secondary CD8 T cell responses, which correlated with decreased IFN-γ production and degranulation by Tim-3 KO cells stimulated with peptide Ag ex vivo. To address the T cell-intrinsic role of Tim-3, we analyzed responses to L. monocytogenes infection by WT and Tim-3 KO TCR-transgenic CD8 T cells following adoptive transfer into a shared WT host. In this setting, the accumulation of CD8 T cells and the generation of cytokine-producing cells were significantly reduced by the lack of Tim-3, demonstrating that this molecule has a direct effect on CD8 T cell function. Combined, our results suggest that Tim-3 can mediate a stimulatory effect on CD8 T cell responses to an acute infection.

SEP-AlgPro: An efficient allergen prediction tool utilizing traditional machine learning and deep learning techniques with protein language model features.

Allergy is a hypersensitive condition in which individuals develop objective symptoms when exposed to harmless substances at a dose that would cause no harm to a "normal" person. Most current computational methods for allergen identification rely on homology or conventional machine learning using limited set of feature descriptors or validation on specific datasets, making them inefficient and inaccurate. Here, we propose SEP-AlgPro for the accurate identification of allergen protein from sequence information. We analyzed 10 conventional protein-based features and 14 different features derived from protein language models to gauge their effectiveness in differentiating allergens from non-allergens using 15 different classifiers. However, the final optimized model employs top 10 feature descriptors with top seven machine learning classifiers. Results show that the features derived from protein language models exhibit superior discriminative capabilities compared to traditional feature sets. This enabled us to select the most discriminatory baseline models, whose predicted outputs were aggregated and used as input to a deep neural network for the final allergen prediction. Extensive case studies showed that SEP-AlgPro outperforms state-of-the-art predictors in accurately identifying allergens. A user-friendly web server was developed and made freely available at https://balalab-skku.org/SEP-AlgPro/, making it a powerful tool for identifying potential allergens.

HOTGpred: Enhancing human O-linked threonine glycosylation prediction using integrated pretrained protein language model-based features and multi-stage feature selection approach.

O-linked glycosylation is a complex post-translational modification (PTM) in human proteins that plays a critical role in regulating various cellular metabolic and signaling pathways. In contrast to N-linked glycosylation, O-linked glycosylation lacks specific sequence features and maintains an unstable core structure. Identifying O-linked threonine glycosylation sites (OTGs) remains challenging, requiring extensive experimental tests. While bioinformatics tools have emerged for predicting OTGs, their reliance on limited conventional features and absence of well-defined feature selection strategies limit their effectiveness. To address these limitations, we introduced HOTGpred (Human O-linked Threonine Glycosylation predictor), employing a multi-stage feature selection process to identify the optimal feature set for accurately identifying OTGs. Initially, we assessed 25 different feature sets derived from various pretrained protein language model (PLM)-based embeddings and conventional feature descriptors using nine classifiers. Subsequently, we integrated the top five embeddings linearly and determined the most effective scoring function for ranking hybrid features, identifying the optimal feature set through a process of sequential forward search. Among the classifiers, the extreme gradient boosting (XGBT)-based model, using the optimal feature set (HOTGpred), achieved 92.03 % accuracy on the training dataset and 88.25 % on the balanced independent dataset. Notably, HOTGpred significantly outperformed the current state-of-the-art methods on both the balanced and imbalanced independent datasets, demonstrating its superior prediction capabilities. Additionally, SHapley Additive exPlanations (SHAP) and ablation analyses were conducted to identify the features contributing most significantly to HOTGpred. Finally, we developed an easy-to-navigate web server, accessible at https://balalab-skku.org/HOTGpred/, to support glycobiologists in their research on glycosylation structure and function.

mACPpred 2.0: Stacked Deep Learning for Anticancer Peptide Prediction with Integrated Spatial and Probabilistic Feature Representations.

Anticancer peptides (ACPs), naturally occurring molecules with remarkable potential to target and kill cancer cells. However, identifying ACPs based solely from their primary amino acid sequences remains a major hurdle in immunoinformatics. In the past, several web-based machine learning (ML) tools have been proposed to assist researchers in identifying potential ACPs for further testing. Notably, our meta-approach method, mACPpred, introduced in 2019, has significantly advanced the field of ACP research. Given the exponential growth in the number of characterized ACPs, there is now a pressing need to create an updated version of mACPpred. To develop mACPpred 2.0, we constructed an up-to-date benchmarking dataset by integrating all publicly available ACP datasets. We employed a large-scale of feature descriptors, encompassing both conventional feature descriptors and advanced pre-trained natural language processing (NLP)-based embeddings. We evaluated their ability to discriminate between ACPs and non-ACPs using eleven different classifiers. Subsequently, we employed a stacked deep learning (SDL) approach, incorporating 1D convolutional neural network (1D CNN) blocks and hybrid features. These features included the top seven performing NLP-based features and 90 probabilistic features, allowing us to identify hidden patterns within these diverse features and improve the accuracy of our ACP prediction model. This is the first study to integrate spatial and probabilistic feature representations for predicting ACPs. Rigorous cross-validation and independent tests conclusively demonstrated that mACPpred 2.0 not only surpassed its predecessor (mACPpred) but also outperformed the existing state-of-the-art predictors, highlighting the importance of advanced feature representation capabilities attained through SDL. To facilitate widespread use and accessibility, we have developed a user-friendly for mACPpred 2.0, available at https://balalab-skku.org/mACPpred2/.