Search Results (126)

Background: Accurately predicting protein druggability is crucial for successful drug development, as it significantly reduces the time and resources required to identify viable drug targets. However, existing methods often face trade-offs between accuracy, efficiency, and interpretability. This study aims to introduce a lightweight framework designed to address these challenges effectively. Methods: We present a lightweight framework that embeds proteins into four biologically informed, non-Euclidean metric spaces, derived from analyses of amino acid sequences, predicted secondary structures, and curated post-translational modification (PTM) annotations. These representations capture key features such as hydrophobicity profiles, PTM densities, spatial patterns, and secondary structure composition, providing interpretable proxies for structure-related determinants of druggability. This approach enhances our understanding of protein functionality while improving druggability predictability in a biologically relevant context. Results: Evaluated on an Aspirin-binding protein dataset using leave-one-out cross-validation (LOOCV), our distance-based ensemble achieves 92.25% accuracy (AUC = 0.9358) in the whole-protein setting. This performance significantly outperforms common sequence-only baselines in the literature while remaining computationally efficient. Conclusions: On a refined single-chain subset, our framework demonstrates performance comparable to established feature engineering pipelines, highlighting its potential effectiveness in practical applications. Together, these results strongly suggest that biologically grounded, non-Euclidean embeddings provide an effective and interpretable alternative to resource-intensive 3D pipelines for target assessment in drug discovery. This approach not only enhances our ability to assess protein druggability but also streamlines the overall process of target identification and validation. Full article

Background/Objectives: Pulmonary hypertension (PH) is a progressive disorder with high morbidity and mortality, partly driven by chronic inflammation. Soluble urokinase plasminogen activator receptor (suPAR) reflects immune activation. We evaluated whether suPAR is altered in Group 1 and Group 4 PH and its association with clinical, echocardiographic, and laboratory parameters. Methods: We enrolled 44 PH patients (36 in Group 1, 8 in Group 4) and 45 healthy controls. All underwent clinical and echocardiographic assessments; right heart catheterization was performed in the PH patients. Serum suPAR was measured by ELISA. N-terminal pro B-type natriuretic peptide (NT-proBNP) and C-reactive protein (CRP) were also assessed. Results: The suPAR plasma levels in the PH group were between 23.91 and 960.8 pg/mL (median: 73.14 p25: 62.77, p75: 167.13). suPAR was significantly higher in PH versus controls (73.14 [62.77–167.13] vs. 65.52 [53.06–80.91] pg/mL; p = 0.012). In logistic regression, systolic blood pressure, erythrocyte sedimentation rate, NT-proBNP, and suPAR independently predicted PH. suPAR correlated negatively with six-minute walk distance (r = −0.310) and tricuspid annular plane systolic excursion (r = −0.295) but positively with systolic pulmonary artery pressure (r = 0.241). On multivariate analysis, six-minute walk distance was the only independent correlate of suPAR (p = 0.004). suPAR levels did not differ between Group 1 and Group 4 PH. Conclusions: suPAR is elevated in Group 1 and Group 4 PH and correlates with functional and echocardiographic indices of disease severity. Larger prospective studies are needed to determine suPAR’s role in diagnosis, risk stratification, and therapeutic decision-making. Full article

Bisphenol AF (BPAF) is widely utilized as an analog of bisphenol A (BPA) in the plastics industry. However, there is limited evidence on its neurodevelopmental toxicity. Existing studies suggest that BPAF has greater accumulation in vivo than other bisphenol analogs, and could pass through the placental barrier and the blood–brain barrier. In this study, we used the human neural progenitor cells line ReNcell CX, which was derived from 14-week human cortical brain tissue, as an in vitro model to investigate the neurodevelopmental toxicity effects of BPAF and BPA on ReNcell CX cells, and explored the possible mechanism by which BPAF induced neurodevelopmental toxicity on ReNcell CX cells. The results showed that BPAF reduced the proliferation of neural progenitor cells and changed the differentiation towards neurons after exposure for 24 h. Compared with BPA, ReNcell CX cells are more susceptible to BPAF exposure. In a 3D neurospheres model, BPAF affected the distance that neurons migrated outwards at the concentration of 2 μM. Furthermore, BPAF increased ROS levels in cells and reduced the expression of key proteins in the Nrf2/HO-1 pathway and its downstream molecules, such as SOD, GSH, and CAT. In conclusion, BPAF induces damage to critical nodes in neural progenitor cell development through the Nrf2/HO-1 pathway. Therefore, clarifying its neurodevelopmental toxicity and elaborating on the neurodevelopmental toxicity effects and mechanisms of bisphenol AF will help identify intervention targets for neurodevelopmental toxicity, and will have important public health significance for the safety assessment and risk prediction of bisphenol-related chemicals. Full article

Objective: Falciparum malaria is a major global health concern, affecting more than half of the world’s population and causing over half a million deaths annually. Red cell invasion is a crucial step in the parasite’s life cycle, where the parasite invade human erythrocytes to sustain infection and ensure survival. Two parasite proteins, Apical Membrane Antigen 1 (AMA-1) and Rhoptry Neck Protein 2 (RON2), are involved in tight junction formation, which is an essential step in parasite invasion of the red blood cell. Targeting the AMA-1 and RON2 interaction with inhibitors halts the formation of the tight junction, thereby preventing parasite invasion, which is detrimental to parasite survival. This study leverages machine learning (ML) to predict potential small molecule inhibitors of the AMA-1–RON2 interaction, providing putative antimalaria compounds for further chemotherapeutic exploration. Method: Data was retrieved from the PubChem database (AID 720542), comprising 364,447 inhibitors and non-inhibitors of the AMA-1–RON2 interaction. The data was processed by computing Morgan fingerprints and divided into training and testing with an 80:20 ratio, and the classes in the training data were balanced using the Synthetic Minority Oversampling Technique. Five ML models developed comprised Random Forest (RF), Gradient Boost Machines (GBMs), CatBoost (CB), AdaBoost (AB) and Support Vector Machine (SVM). The performances of the models were evaluated using accuracy, F1 score, and receiver operating characteristic—area under the curve (ROC-AUC) and validated using held-out data and a y-randomization test. An applicability domain analysis was carried out using the Tanimoto distance with a threshold set at 0.04 to ascertain the sample space where the models predict with confidence. Results: The GBMs model emerged as the best, achieving 89% accuracy and a ROC-AUC of 92%. CB and RF had accuracies of 88% and 87%, and ROC-AUC scores of 93% and 91%, respectively. Conclusions: Experimentally validated inhibitors of the AMA-1–RON2 interaction could serve as starting blocks for the next-generation antimalarial drugs. The models were deployed as a web-based application, known as PLASMOpred. Full article

Protein content is a key quality indicator in nuts, influencing their color, taste, storage, and processing properties. Traditional methods for protein quantification, such as the Kjeldahl nitrogen method, are time-consuming and destructive, highlighting the need for rapid, convenient alternatives. This study explores the feasibility of using portable near-infrared spectroscopy (NIRS) for the quantitative prediction of protein content in Torreya grandis (T. grandis) kernels by comparing different sample states (with shell, without shell, and granules). Spectral data were acquired using a portable NIR spectrometer, and the protein content was determined via the Kjeldahl nitrogen method as a reference. Outlier detection was performed using principal component analysis combined with Mahalanobis distance (PCA-MD) and concentration residual analysis. Various spectral preprocessing techniques and partial least squares regression (PLSR) were applied to develop protein prediction models. The results demonstrated that portable NIRS could effectively predict protein content in T. grandis kernels, with the best performance being achieved using granulated samples. The optimized model (1Der-SNV-PLSR-G) significantly outperformed models based on whole kernels (with or without shell), with determination coefficients for the calibration set ($R_c^2$) and prediction set ($R_p^2$) of 0.92 and 0.86, respectively, indicating that the sample state critically influenced prediction accuracy. This study confirmed the potential of portable NIRS as a rapid and convenient tool for protein quantification in nuts, offering a practical alternative to conventional methods. The findings also suggested its broader applicability for quality assessment in other nuts and food products, contributing to advancements in food science and agricultural technology. Full article

The poultry industry plays a pivotal role in global agriculture, with poultry serving as a major source of protein and contributing significantly to economic growth. However, the sector faces challenges associated with labor-intensive tasks that are repetitive and physically demanding. Automation has emerged as a critical solution to enhance operational efficiency and improve working conditions. Specifically, robotic manipulation and handling of objects is becoming ubiquitous in factories. However, challenges exist to precisely identify and guide a robot to handle a pile of objects with similar textures and colors. This paper focuses on the development of a vision system for a robotic solution aimed at automating the chicken rehanging process, a fundamental yet physically strenuous activity in poultry processing. To address the limitation of the generic instance segmentation model in identifying overlapped objects, a cost-effective, dual-active laser scanning system was developed to generate precise depth data on objects. The well-registered depth data generated were integrated with the RGB images and sent to the instance segmentation model for individual chicken detection and identification. This enhanced approach significantly improved the model’s performance in handling complex scenarios involving overlapping chickens. Specifically, the integration of RGB-D data increased the model’s mean average precision (mAP) detection accuracy by 4.9% and significantly improved the center offset—a customized metric introduced in this study to quantify the distance between the ground truth mask center and the predicted mask center. Precise center detection is crucial for the development of future robotic control solutions, as it ensures accurate grasping during the chicken rehanging process. The center offset was reduced from 22.09 pixels (7.30 mm) to 8.09 pixels (2.65 mm), demonstrating the approach’s effectiveness in mitigating occlusion challenges and enhancing the reliability of the vision system. Full article

Background/Objectives: RNA research is critical for understanding gene regulation, disease mechanisms, and therapeutic development. Constructing effective RNA benchmark models for accurate downstream analysis has become a significant research challenge. The objective of this study is to propose a robust benchmark model, DRFormer, for RNA sequence downstream tasks. Methods: The DRFormer model utilizes RNA sequences to construct novel vision features based on secondary structure and sequence distance. These features are pre-trained using the SWIN model to develop a SWIN-RNA submodel. This submodel is then integrated with an RNA sequence model to construct a multimodal model for downstream analysis. Results: We conducted experiments on various RNA downstream tasks. In the sequence classification task, the MCC reached 94.4%, surpassing the state-of-the-art RNAErnie model by 1.2%. In the protein–RNA interaction prediction, DRFormer achieved an MCC of 0.492, outperforming advanced models like BERT-RBP and PrismNet. In RNA secondary structure prediction, the F1 score was 0.690, exceeding the widely used SPOT-RNA model by 1%. Additionally, generalization experiments on DNA tasks yielded satisfactory results. Conclusions: DRFormer is the first RNA sequence downstream analysis model that leverages structural features to construct a vision model and integrates sequence and vision models in a multimodal manner. This approach yields excellent prediction and analysis results, making it a valuable contribution to RNA research. Full article

Background/Objectives: The normal cellular prion protein (PrP^C) is a cell-surface glycoprotein, mainly localised in neurons of the central nervous system (CNS). The human PRNP gene encodes 253 amino acid residues of precursor PrP^C. Several studies that investigated the role of PRNP and PrP^C in placental mammals, such as humans and mice, failed to reveal its exact function. Methods: In this study, we sequenced and characterised the PRNP gene and PrP^C of the marsupial, P. calura, as a strategy to gain molecular insights into its structure and physicochemical properties. Placentals are separated from marsupials by approximately 125 million years of independent evolution. Results: Standard Western blotting analysis of PrP^C phascogale displayed the typical un-, mono-, and di-glycosylated bands recognized in placentals. Furthermore, we showed that phascogale PRNP gene has two exons, similar to all the marsupials and placentals of the PRNP genes studied. Of note, the phascogale PRNP gene contained distinctive repeats in the PrP^C tail region comparable to the closely related Tasmanian devil (Sarcophilus harrisii) and more distantly related to the grey short-tailed opossum (Monodelphis domestica), common wombat (Vombatus ursinus), and Tammar wallaby (Macropus eugenii); however, its specific composition and numbers were different from placentals. Of importance, comparisons of the phascogale’s PrP^C physicochemical properties with other monotremes, marsupials, and placentals confirmed the Monotremata–Marsupialia–Placentalia evolutionary distance. We found that the protein instability index, a method used to predict the stability of a protein in vivo (Stable: <40; Instable >40), showed that the PrP^C of all marsupials tested, including phascogale, were highly stable compared with the birds, reptiles, amphibians, and fish that were shown to be highly unstable. However, the instability index predicted that all placental species, including human (Homo sapiens), mouse (Mus musculus), bank vole (Myodes glareolus), rhinoceros (Rhinocerotidae), dog (Canis lupus familiaris), flying fox (Pteropus vampyrus), whale (Physeter catodon), cattle (Bos taurus), and sheep (Ovis aries), were either slightly unstable or nearly unstable. Further, our analysis revealed that despite their predicted high PrP^C stability, P. calura exhibited substantial N-terminal disorder (53.76%), while species with highly unstable PrP^Cs based on their instability index, such as Danio rerio, Oryzias latipes, and Astyanax mexicanus, displayed even higher levels of N-terminal disorder (up to 75.84%). These findings highlight a discrepancy between overall predicted stability and N-terminal disorder, suggesting a potential compensatory role of disorder in modulating prion protein stability and function. Conclusions: These results suggest that the high stability of marsupial prion proteins indicates a vital role in maintaining protein homeostasis; however more work is warranted to further depict the exact function. Full article

First believed to be a simple intermediary between the information encoded in deoxyribonucleic acid and that functionally displayed in proteins, ribonucleic acid (RNA) is now known to have many functions through its abundance and intricate, ubiquitous, diverse, and dynamic structure. About 70–90% of the human genome is transcribed into protein-coding and noncoding RNAs as main determinants along with regulatory sequences of cellular to populational biological diversity. From the nucleotide sequence or primary structure, through Watson–Crick pairing self-folding or secondary structure, to compaction via longer distance Watson–Crick and non-Watson–Crick interactions or tertiary structure, and interactions with RNA or other biopolymers or quaternary structure, or with metabolites and biomolecules or quinary structure, RNA structure plays a critical role in RNA’s lifecycle from transcription to decay and many cellular processes. In contrast to the success of 3-dimensional protein structure prediction using AlphaFold, RNA tertiary and beyond structures prediction remains challenging. However, approaches involving machine learning and artificial intelligence, sequencing of RNA and its modifications, and structural analyses at the single-cell and intact tissue levels, among others, provide an optimistic outlook for the continued development and refinement of RNA-based applications. Here, we highlight those in gene therapy. Full article

Living systems cannot rely on random intermolecular approaches toward cell crowding, and hidden mechanisms must be present to favor only those molecular interactions required explicitly by the biological function. Electromagnetic messaging among proteins is proposed from the observation that charged amino acids located on the protein surface are mostly in adjacent sequence positions and/or in spatial proximity. Molecular dynamics (MD) simulations have been used to predict electric charge proximities arising from concerted motions of charged amino acid side chains in two protein model systems, human ubiquitin and the chitinolytic enzyme from Ostrinia furnacalis. This choice has been made for their large difference in size and sociality. Protein electrodynamics seems to emerge as the framework for a deeper understanding of the long-distance interactions of proteins with their molecular environment. Our findings will be valuable in orienting the design of proteins with specific recognition patterns. Full article

The specific objective of the present study was to develop computational models, by means of which predictions could be performed regarding the quality of the bulk-tank milk in dairy sheep and goat farms. Our hypothesis was that use of specific variables related to the health management applied in the farm can facilitate the development of predictions regarding values related to milk quality, specifically for fat content, protein content, fat and protein content combined, somatic cell counts, and total bacterial counts. Bulk-tank milk from 325 sheep and 119 goat farms was collected and evaluated by established techniques for analysis of fat and protein content, for somatic cell counting, and for total bacterial counting. Subsequently, computational models were constructed for the prediction of five target values: (a) fat content, (b) protein content, (c) fat and protein, (d) somatic cell counts, and (e) total bacterial counts, through the use of 21 independent variables related to factors prevalent in the farm. Five machine learning tools were employed: decision trees (18 different models evaluated), random forests (16 models), XGBoost (240 models), k-nearest neighbours (72 models), and neural networks (576 models) (in total, 9220 evaluations were performed). Tools found with the lowest mean absolute percentage error (MAPE) between the five tools used to test predictions for each target value were selected. In sheep farms, for the prediction of protein content, k-nearest neighbours was selected (MAPE: 3.95%); for the prediction of fat and protein content combined, neural networks was selected (6.00%); and for the prediction of somatic cell counts, random forests and k-nearest neighbours were selected (6.55%); no tool provided useful predictions for fat content and for total bacterial counts. In goat farms, for the prediction of protein content, k-nearest neighbours was selected (MAPE: 6.17%); for the prediction of somatic cell counts, random forests and k-nearest neighbours were selected (4.93% and 5.00%); and for the prediction of total bacterial counts, neural networks was selected (8.33%); no tool provided useful prediction models for fat content and for fat and protein content combined. The results of the study will be of interest to farmers, as well as to professionals; the findings will also be useful to dairy processing factories. That way, it will be possible to obtain a distance-aware, rapid, quantitative estimation of the milk output from sheep and goat farms with sufficient data attributes. It will thus become easier to monitor and improve milk quality at the farm level as part of the dairy production chain. Moreover, the findings can support the setup of relevant and appropriate measures and interventions in dairy sheep and goat farms. Full article

For the first time, we describe phylogenomic signatures of an epidemic lineage of vesicular stomatitis Indiana virus (VSIV). We applied multiple evolutionary analyses to a dataset of 87 full-length genome sequences representing the circulation of an epidemic VSIV lineage in the US between 2019 and 2020. Based on phylogenetic analyses, we predicted the ancestral relationship of this lineage with a specific group of isolates circulating in the endemic zone of Chiapas, Mexico. Subsequently, our findings indicate that the lineage diversified into at least four different subpopulations during its circulation in the US. We identified single nucleotide polymorphisms (SNPs) that differentiate viral subpopulations and assessed their potential relevance using comparative phylogenetic methods, highlighting the preponderance of synonymous mutations during the differentiation of these populations. Purifying selection was the main evolutionary force favoring the conservation of this epidemic phenotype, with P and G genes as the main drivers of the evolution of this lineage. Our analyses identified multiple codon sites under positive selection and the association of these sites with specific functional domains at P, M, G, and L proteins. Based on ancestral reconstruction analyses, we showed the potential relevance of some of the sites identified under positive selection to the adaptation of the epidemic lineage at the population level. Finally, using a representative group of viruses from Colorado, we established a positive correlation between genetic and geographical distances, suggesting that positive selection on specific codon positions might have favored the adaptation of different subpopulations to circulation in specific geographical settings. Collectively, our study reveals the complex dynamics that accompany the evolution of an epidemic lineage of VSIV in nature. Our analytical framework provides a model for conducting future evolutionary analyses. The ultimate goal is to support the implementation of an early warning system for vesicular stomatitis virus in the US, enabling early detection of epidemic precursors from Mexico. Full article

The genus Pourthiaea Decne., a deciduous woody group with high ornamental value, belongs to the family Rosaceae. Here, we reported newly sequenced plastid genome sequences of Pourthiaea beauverdiana (C. K. Schneid.) Hatus., Pourthiaea parvifolia E. Pritz., Pourthiaea villosa (Thunb.) Decne., and Photinia glomerata Rehder & E. H. Wilson. The plastomes of these three Pourthiaea species shared the typical quadripartite structures, ranging in size from 159,903 bp (P. parvifolia) to 160,090 bp (P. beauverdiana). The three Pourthiaea plastomes contained a pair of inverted repeat regions (26,394–26,399 bp), separated by a small single-copy region (19,304–19,322 bp) and a large single-copy region (87,811–87,973 bp). A total of 113 unique genes were predicted for the three Pourthiaea plastomes, including four ribosomal RNA genes, 30 transfer RNA genes, and 79 protein-coding genes. Analyses of inverted repeat/single-copy boundary, mVISTA, nucleotide diversity, and genetic distance showed that the plastomes of 13 Pourthiaea species (including 10 published plastomes) are highly conserved. The number of simple sequence repeats and long repeat sequences is similar among 13 Pourthiaea species. The three non-coding regions (trnT-GGU-psbD, trnR-UCU-atpA, and trnH-GUG-psbA) were the most divergent. Only one plastid protein-coding gene, rbcL, was under positive selection. Phylogenetic analyses based on 78 shared plastid protein-coding sequences and 29 nrDNA sequences strongly supported the monophyly of Pourthiaea. As for the relationship with other genera in our phylogenies, Pourthiaea was sister to Malus in plastome phylogenies, while it was sister to the remaining genera in nrDNA phylogenies. Furthermore, significant cytonuclear discordance likely stems from hybridization events within Pourthiaea, reflecting complex evolutionary dynamics within the genus. Our study provides valuable genetic insights for further phylogenetic, taxonomic, and species delimitation studies in Pourthiaea, as well as essential support for horticultural improvement and conservation of the germplasm resources. Full article

Predicting protein–ligand binding sites is an integral part of structural biology and drug design. A comprehensive understanding of these binding sites is essential for advancing drug innovation, elucidating mechanisms of biological function, and exploring the nature of disease. However, accurately identifying protein–ligand binding sites remains a challenging task. To address this, we propose PGpocket, a geometric deep learning-based framework to improve protein–ligand binding site prediction. Initially, the protein surface is converted into a point cloud, and then the geometric and chemical properties of each point are calculated. Subsequently, the point cloud graph is constructed based on the inter-point distances, and the point cloud graph neural network (GNN) is applied to extract and analyze the protein surface information to predict potential binding sites. PGpocket is trained on the scPDB dataset, and its performance is verified on two independent test sets, Coach420 and HOLO4K. The results show that PGpocket achieves a 58% success rate on the Coach420 dataset and a 56% success rate on the HOLO4K dataset. These results surpass competing algorithms, demonstrating PGpocket’s advancement and practicality for protein–ligand binding site prediction. Full article

The human parainfluenza virus type 4 (HPIV4) can be classified into two distinct subtypes, 4a and 4b. The full lengths of the fusion gene (F gene) of 48 HPIV4 strains collected during the period of 1966–2022 were analyzed. Based on these gene sequences, the time-scaled evolutionary tree was constructed using Bayesian Markov chain Monte Carlo methods. A phylogenetic tree showed that the first division of the two subtypes occurred around 1823, and the most recent common ancestors of each type, 4a and 4b, existed until about 1940 and 1939, respectively. Although the mean genetic distances of all strains were relatively wide, the distances in each subtype were not wide, indicating that this gene was conserved in each subtype. The evolutionary rates of the genes were relatively low (4.41 × 10⁻⁴ substitutions/site/year). Moreover, conformational B-cell epitopes were predicted in the apex of the trimer fusion protein. These results suggest that HPIV4 subtypes diverged 200 years ago and the progenies further diverged and evolved. Full article