pLM4Alg: Protein Language Model-Based Predictors for Allergenic Proteins and Peptides

Journal of Agricultural and Food Chemistry, Journal Year: 2023, Volume and Issue: 72(1), P. 752 - 760

Published: Dec. 19, 2023

The rising prevalence of allergy demands efficient and accurate bioinformatic tools to expedite allergen identification risk assessment while also reducing wet experiment expenses time. Recently, pretrained protein language models (pLMs) have successfully predicted structure function. However, our best knowledge, they not been used for predicting allergenic proteins/peptides. Therefore, this study aims develop robust protein/peptide prediction using five pLMs varying sizes systematically assess their performance through fine-tuning with a convolutional neural network. developed pLM4Alg achieved state-of-the-art accuracy, Matthews correlation coefficient, area under the curve scoring 93.4–95.1%, 0.869–0.902, 0.981–0.990, respectively. Moreover, is first model capable handling tasks involving residue-missed sequences containing nonstandard amino acid residues. To facilitate easy access, user-friendly web server (https://f6wxpfd3sh.us-east-1.awsapprunner.com) has established. expected become leading machine learning-based peptides proteins. Its collaboration other predictors holds great promise accelerating research.

Language: Английский

---

DeepAlgPro: an interpretable deep neural network model for predicting allergenic proteins

Briefings in Bioinformatics, Journal Year: 2023, Volume and Issue: 24(4)

Published: June 14, 2023

Allergies have become an emerging public health problem worldwide. The most effective way to prevent allergies is find the causative allergen at source and avoid re-exposure. However, of current computational methods used identify allergens were based on homology or conventional machine learning methods, which inefficient still had room be improved for detection with low homology. In addition, few deep reported, although has been successfully applied several tasks in protein sequence analysis. present work, a neural network-based model, called DeepAlgPro, was proposed allergens. We showed its great accuracy applicability large-scale forecasts by comparing it other available tools. Additionally, we ablation experiments demonstrate critical importance convolutional module our model. Moreover, further analyses that epitope features contributed model decision-making, thus improving model's interpretability. Finally, found DeepAlgPro capable detecting potential new Overall, can serve as powerful software identifying

Language: Английский

---

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

International Journal of Biological Macromolecules, Journal Year: 2024, Volume and Issue: 273, P. 133085 - 133085

Published: June 12, 2024

Language: Английский

---

Comparative Analysis of pK_a Predictions for Arsonic Acids Using Density Functional Theory-Based and Machine Learning Approaches

ACS Omega, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 16, 2025

Arsonic acids (RAsO(OH)2), prevalent in contaminated food, water, air, and soil, pose significant environmental health risks due to their variable ionization states, which influence key properties such as lipophilicity, solubility, membrane permeability. Accurate pK a prediction for these compounds is critical yet challenging, existing models often exhibit limitations across diverse chemical spaces. This study presents comparative analysis of predictions arsonic using support vector machine-based machine learning (ML) approach three density functional theory (DFT)-based models. The DFT evaluated include correlations the maximum surface electrostatic potential (V S,max), atomic charges derived from solvation model (solvation based on density), scaled solvent-accessible method. Results indicate that yielded high mean unsigned errors, rendering it less effective. In contrast, charge-based method conjugated arsonate base provided most accurate predictions. ML-based demonstrated strong predictive performance, suggesting its utility broader obtained values V S,max show weak level, because way predicting related only character molecule. However, influenced by many factors, including molecular structure, solvation, resonance, inductive effects, local environments. cannot fully capture different interactions, gives simplistic view overall field.

Language: Английский

---

Prospects for developing allergen‐depleted food crops

The Plant Genome, Journal Year: 2023, Volume and Issue: 16(4)

Published: Aug. 28, 2023

In addition to the challenge of meeting global demand for food production, there are increasing concerns about safety and need protect consumer health from negative effects foodborne allergies. Certain bio-molecules (usually proteins) present in can act as allergens that trigger unusual immunological reactions, with potentially life-threatening consequences. The relentless working lifestyles modern era often incorporate poor eating habits include readymade prepackaged processed foods, which contain additives such peanuts, tree nuts, wheat, soy-based products, rather than traditional home cooking. Of predominant allergenic foods (soybean, fish, peanut, shellfish, eggs, milk), peanuts (Arachis hypogaea) best characterized source allergens, followed by nuts (Juglans regia, Prunus amygdalus, Corylus avellana, Carya illinoinensis, Anacardium occidentale, Pistacia vera, Bertholletia excels), wheat (Triticum aestivum), soybeans (Glycine max), kidney beans (Phaseolus vulgaris). prevalence allergies has risen significantly recent years including chance accidental exposure foods. contrast, standards detection, diagnosis, cure have not kept pace unfortunately suboptimal. this review, we mainly focus on associated soybean, bean, highlighting their physiological properties functions well considering research directions tailoring allergen gene expression. particular, discuss how advances molecular breeding, genetic engineering, genome editing be used develop potential low crops health.

Language: Английский

---

Accelerating the identification of the allergenic potential of plant proteins using a stacked ensemble-learning framework

Journal of Biomolecular Structure and Dynamics, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 13

Published: Feb. 22, 2024

Plant-allergenic proteins (PAPs) have the potential to induce allergic reactions in certain individuals. While these are generally innocuous for majority of people, they can elicit an immune response those with particular sensitivities. Thus, screening and prioritizing allergenic plant is indispensable development diagnostic tools, therapeutic interventions or medications treat reactions. However, investigating based on experimental methods costly labour-intensive. Therefore, we develop StackPAP, a three-layer stacking ensemble framework accurate large-scale identification PAPs. In at first layer, conducted comprehensive analysis extensive set feature descriptors. Subsequently, selected fused five sequence-based descriptors, including amphiphilic pseudo-amino acid composition, dipeptide deviation from expected mean, amino pseudo composition composition. Additionally, applied efficient genetic algorithm (GA-SAR) determine informative sets. second 12 powerful machine learning (ML) methods, combination all sets, were employed construct pool base classifiers. Finally, 13 classifiers using GA-SAR method combined final meta-classifier. Our results revealed promising prediction performance accuracy, Matthew's correlation coefficient AUC 0.984, 0.969 0.993, respectively, as judged by independent test dataset. conclusion, both cross-validation indicated superior StackPAP compared several ML-based To accelerate allergenicity proteins, developed user-friendly web server (https://pmlabqsar.pythonanywhere.com/StackPAP). We anticipate that will be useful tool rapidly PAPs vast number proteins.

Language: Английский

---

Seed storage allergens tackled via next-generation research assistant

Frontiers in Food Science and Technology, Journal Year: 2024, Volume and Issue: 4

Published: March 12, 2024

The expanding consumption of plant proteins in the diet to overcome environmental issues associated with animal is increasing incidence food-induced allergic reactions. One 21st-century research drivers agriculture sciences development and validation concrete approaches for modulating expression allergenic crops before harvesting. food allergies primarily induced by seed storage that clinicians are experiencing recently because more predominant use plant-derived industry. Increased availability high-throughput technologies has generated an ever-growing number omics data, allowing us have better structural knowledge SSPs molecular properties can inform allergenicity assessment. recent systems targeted genome engineering, without double-strand DNA breaks, allow introduction precise modifications directly into commercial species. Artificial intelligence significantly transforming scientific across every stage, assisting scientists, processing large-scale making predictions, automating tasks. During this epochal change, marked encounter between artificial synthetic biology, a next-generation assistant (NGA) coming alive. Here, we propose new conceptual vision facilitate speed up editing cross-reactivity sites obtain hypoallergenic cultivars avoid pleiotropic effects. Finally, discuss potential applications way conceive research. NGA may be undoubtedly capable managing evolution SPP through prediction novel epitopes, as well immunological response mechanisms.

Language: Английский

---

Role of Bioinformatics Tools in the Food Processing Industry

Published: April 25, 2024

The food processing industry plays a significant role in providing safe and nutritious to the growing population. examination of ingredients, detection potential pollutants, knowledge on microbiome facilities can all be aided by bioinformatics technologies. purpose this chapter is describe tools industry. It emphasizes metagenomics transcriptomics understanding microbial diversity dynamics environments. Additionally, it highlights use predicting allergenicity developing personalized nutrition. also explores applications fraud traceability. adoption enhance safety, improve quality control, reduce risk foodborne illnesses. However, implementation these requires specialized skills, infrastructure, resources. This concludes discussing challenges opportunities associated with integration widespread promote data-driven evidence-based approach processing, which ultimately lead production safer healthier products.

Language: Английский

---

A deep learning based ensemble approach for protein allergen classification

PeerJ Computer Science, Journal Year: 2023, Volume and Issue: 9, P. e1622 - e1622

Published: Oct. 12, 2023

In recent years, the increased population has led to an increase in demand for various industrially processed edibles and other consumable products. These industries regularly alter proteins found raw materials generate more commercially viable end-products order keep up with consumer demand. modifications result a substance that may cause allergic reactions consumers, thereby creating protein allergen. The detection of such substances is essential prevention, diagnosis treatment conditions. Bioinformatics computational methods can be used analyze information contained amino-acid sequences detect possible allergens. article presents deep learning based ensemble approach identify allergens using Extra Tree, Deep Belief Network (DBN), CatBoost models. proposed model achieves higher accuracy by combining prediction results three models majority voting. evaluation was carried out on benchmark allergen dataset, performance analysis revealed outperforms state-of-the-art literature techniques 89.16%.

Language: Английский

---

A New Hybrid Classification Framework in Childhoods Allergies with Dataset Slicing Method

Düzce Üniversitesi Bilim ve Teknoloji Dergisi, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 11, 2024

Childhood allergies, particularly food are growing more frequent. Their major influence on children's health and well-being has piqued the interest of worldwide public officials. The increased prevalence childhood allergies in Turkey, where these patterns also relevant, adds urgency to need for effective classification management options. This study addresses shortcomings simple algorithms obtaining high accuracy by presenting a novel hybrid methodology. research creates method three different prediction models built combining Support Vector Machine Decision Tree classifiers. improves process taking into account instances that have been incorrectly classified as possible sources useful information instead just being noise. instance filtering-based algorithm is used this maintains simplicity interpreting learning outcomes while achieving comparatively accuracy. Extensive experiments allergy dataset show effectiveness approach, with an impressive 0.906. greatly outperforms fundamental algorithms. experimental outputs important implications medical professionals. might add valuable contribution literature giving fresh solution classification.

Language: Английский

---