Computational prediction of allergenic proteins based on multi-feature fusion

Article Biochemistry & Molecular Biology

iRNA-ac4C: A novel computational method for effectively detecting N4-acetylcytidine sites in human mRNA

Wei Su et al.

Summary: A novel predictor, iRNA-ac4C, was developed to identify ac4C sites in human mRNA using three feature extraction methods. The results showed promising generalization capabilities.

INTERNATIONAL JOURNAL OF BIOLOGICAL MACROMOLECULES (2023)

Add to Collection

Article Biochemistry & Molecular Biology

AcrPred: A hybrid optimization with enumerated machine learning algorithm to predict Anti-CRISPR proteins

Fu-Ying Dao et al.

Summary: CRISPR-Cas has attracted extensive attention as a gene editing tool. Anti-CRISPR (Acr) proteins can inhibit the CRISPR-Cas defense system and be utilized for gene editing regulation. The study developed a high-accuracy prediction model called AcrPred, based on a two-step model fusion strategy, which achieved an AUC of 0.952 with independent dataset validation. The model demonstrated strong generalization ability by correctly identifying 9 out of 10 new Acr proteins compared to published tools. Additionally, a user-friendly web-server AcrPred was established for easy identification of potential Anti-CRISPR proteins.

INTERNATIONAL JOURNAL OF BIOLOGICAL MACROMOLECULES (2023)

Add to Collection

Article Medicine, General & Internal

Bitter-RF: A random forest machine model for recognizing bitter peptides

Yu-Fei Zhang et al.

Summary: In this study, a Random forest-based model called Bitter-RF was developed to classify bitter peptides more accurately. The model achieved better results than the previous generation model and showed potential for practical applications in bitter peptide research.

FRONTIERS IN MEDICINE (2023)

Add to Collection

Article Biology

m5U-SVM: identification of RNA 5-methyluridine modification sites based on multi-view features of physicochemical features and distributed representation

Chunyan Ao et al.

Summary: In this study, a novel predictor called m5U-SVM was developed to identify m5U modification sites from RNA sequences using multi-view features and machine learning algorithms. The optimized multi-view features were obtained from traditional physicochemical features and distributed representation features. The proposed model outperformed the existing state-of-the-art tool in terms of performance.

BMC BIOLOGY (2023)

Add to Collection

Article Biology

Deciphering the immune heterogeneity dominated by natural killer cells with prognostic and therapeutic implications in hepatocellular carcinoma

Chengbin Guo et al.

Summary: This study identified 80 prognosis-related natural killer (NK) cell marker genes (NKGs) using single-cell RNA-sequencing analysis, and categorized hepatocellular carcinoma (HCC) patients into two subtypes based on these genes. A five-gene prognostic signature-NKscore (UBB, CIRBP, GZMH, NUDC, and NCL) was established, and differences in mutation status and immunotherapy sensitivity between the two NKscore risk groups were revealed. These findings provide a novel NK cell-related signature for predicting prognosis and immunotherapy efficacy in HCC patients.

COMPUTERS IN BIOLOGY AND MEDICINE (2023)

Add to Collection

Editorial Material Biochemical Research Methods

Explainable Artificial Intelligence for Protein Function Prediction: A Perspective View

Nguyen Quoc Khanh Le

CURRENT BIOINFORMATICS (2023)

Add to Collection

Article Biochemical Research Methods

Characterization, Potential Prognostic Value, and Immune Heterogeneity of Cathepsin C in Diffuse Glioma

Quanwei Zhou et al.

Summary: In this study, the performance of CTSC in predicting prognosis and therapeutic targets in diffuse glioma was investigated. The expression profile of CTSC in various tumors and glioma samples was collected. CTSC was found to be aberrantly expressed and significantly correlated with clinical outcomes. It was also associated with immune scores, stromal scores, and infiltrating levels of specific immune cells. Additionally, CTSC was closely correlated with immune checkpoint molecules. These findings indicate that CTSC could serve as an independent indicator of poor prognosis in diffuse glioma and be a potential target for therapy.

CURRENT BIOINFORMATICS (2023)

Add to Collection

Article Biology

A random forest-based metabolic risk model to assess the prognosis and metabolism-related drug targets in ovarian cancer

Haoxin Zhang et al.

Summary: This study identified 17 metabolic pathways with prognostic values in ovarian cancer using data integration and the random forest algorithm. It developed a metabolic risk scoring model and classified patients into two subtypes. The study found differences in prognosis, gene expression, immune signature enrichment, Hallmark signature enrichment, and somatic mutations between the subtypes. It also successfully predicted sensitivity to immunotherapy and chemotherapy drugs and identified drug targets associated with different risk phenotypes.

COMPUTERS IN BIOLOGY AND MEDICINE (2023)

Add to Collection

Article Biology

Identification of SH2 domain-containing proteins and motifs prediction by a deep learning method

Duanzhi Wu et al.

Summary: In this study, SH2 domain-containing proteins and non-SH2 domain-containing proteins were successfully identified using deep learning technology. The best performing 288-dimensional features were obtained. Additionally, a new motif, YKIR, in the SH2 domain was discovered and its function in signal transduction was analyzed.

COMPUTERS IN BIOLOGY AND MEDICINE (2023)

Add to Collection

Article Medicine, General & Internal

A First Computational Frame for Recognizing Heparin-Binding Protein

Wen Zhu et al.

Summary: This study provides the first recognition framework for accurately identifying HBP based on machine learning. By using four sequence descriptors, HBP and non-HBP samples were represented by discrete numbers and input into SVM and RF algorithms for comparison. The SVM-based classifier was found to have the greatest potential for identifying HBP.

DIAGNOSTICS (2023)

Add to Collection

Article Health Care Sciences & Services

A gender specific risk assessment of coronary heart disease based on physical examination data

Hui Yang et al.

Summary: In this study, a gender-specific cascading system for risk assessment of coronary heart disease (CHD) was developed based on physical examination data. A CHD risk model was constructed using a fully connected network (FCN) and a CHD risk scorecard was established using logistic regression (LR) to enhance convenience and flexibility. An online CHD risk assessment system has been established for promoting CHD personal lifestyle management.

NPJ DIGITAL MEDICINE (2023)

Add to Collection

Review Biochemistry & Molecular Biology

Pathogenesis of allergic diseases and implications for therapeutic interventions

Ji Wang et al.

Summary: Allergic diseases, such as allergic rhinitis, allergic asthma, atopic dermatitis, food allergy, and eczema, are systemic diseases caused by immune system dysfunction. The increasing incidence rates of these diseases, along with their high recurrence rates, have attracted significant attention. Their pathogenesis is complex and involves factors such as maternal-fetal environment, living environment, genetics, epigenetics, and immune status. Understanding the influencing factors, pathogenesis, and treatment progress of allergic diseases is becoming increasingly important for doctors and scientists.

SIGNAL TRANSDUCTION AND TARGETED THERAPY (2023)

Add to Collection

Review Biochemical Research Methods

Comparative analysis of machine learning-based approaches for identifying therapeutic peptides targeting SARS-CoV-2

Balachandran Manavalan et al.

Summary: This study comprehensively evaluates the existing IL-6 and AVP prediction algorithms and discusses their advantages and disadvantages. The results provide guidance for the rapid design and development of accurate and efficient computational tools against SARS-CoV-2.

BRIEFINGS IN BIOINFORMATICS (2022)

Add to Collection

Article Computer Science, Information Systems

Towards a better prediction of subcellular location of long non-coding RNA

Zhao-Yue Zhang et al.

Summary: This study presents a support vector machine-based approach that incorporates mutual information algorithm and incremental feature selection strategy to improve the prediction performance of lncRNA subcellular localization.

FRONTIERS OF COMPUTER SCIENCE (2022)

Add to Collection

Article Biochemical Research Methods

A Novel Feature Selection Method Based on MRMR and Enhanced Flower Pollination Algorithm for High Dimensional Biomedical Data

Chaokun Yan et al.

Summary: In this study, a novel feature selection method called MRMR-EFPATS is proposed, which combines MRMR and an improved FPA method. By quickly screening important features and achieving fast convergence to find the optimal subset, it achieves good accuracy and speed on high-dimensional biomedical datasets.

CURRENT BIOINFORMATICS (2022)

Add to Collection

Article Biochemical Research Methods

A Novel Method for Predicting Essential Proteins by Integrating Multidimensional Biological Attribute Information and Topological Properties

Hanyu Lu et al.

Summary: This paper proposes a new method (EOP) for inferring potential essential proteins by combining the multidimensional biological attribute information of proteins with the topological properties of the protein-protein interaction network. The simulation results show that this method identifies more essential proteins compared to other methods, and has higher recognition rates under various conditions.

CURRENT BIOINFORMATICS (2022)

Add to Collection

Article Biochemistry & Molecular Biology

THRONE: A New Approach for Accurate Prediction of Human RNA N7-Methyl-guanosine Sites

Watshara Shoombuatong et al.

Summary: In this study, a novel predictor called THRONE was developed to accurately identify m7G sites in the human genome. THRONE utilizes multiple sequence-based features and machine learning classifiers, and combines multiple models through ensemble learning. The proposed method outperformed existing methods in predicting m7G sites.

JOURNAL OF MOLECULAR BIOLOGY (2022)

Add to Collection

Article Biotechnology & Applied Microbiology

Deepm5C: A deep-learning-based hybrid framework for identifying human RNA N5-methylcytosine sites using a stacking strategy

Md Mehedi Hasan et al.

Summary: The study proposes a new bioinformatics method Deepm5C for identifying RNA m5C sites in the human genome, which achieved more accurate and stable performance than existing predictors and is expected to assist community-wide efforts.

MOLECULAR THERAPY (2022)

Add to Collection

Article Biochemistry & Molecular Biology

AllerCatPro 2.0: a web server for predicting protein allergenicity potential

Minh N. Nguyen et al.

Summary: AllerCatPro 2.0 is a web server for predicting the allergenic potential of proteins with better accuracy than other methods, and new features to help assessors make informed decisions. It predicts protein similarity using amino acid sequences and predicted 3D structures.

NUCLEIC ACIDS RESEARCH (2022)

Add to Collection

Article Biochemical Research Methods

TACOS: a novel approach for accurate prediction of cell-specific long noncoding RNAs subcellular localization

Young-Jun Jeon et al.

Summary: This study presents the first application of the TACOS method to identify the subcellular localization of human lncRNA in 10 different cell types, with comprehensive evaluations and consistent performance compared to other methods.

BRIEFINGS IN BIOINFORMATICS (2022)

Add to Collection

Article Biology

HGSORF: Henry Gas Solubility Optimization-based Random Forest for C-Section prediction and XAI-based cause analysis

Md Saiful Islam et al.

Summary: A stable predictive model is crucial for accurately forecasting cesarean delivery. To improve the accuracy of prediction, a Henry gas solubility optimization-based random forest model has been proposed. The model achieved superior performance and was explained using explainable artificial intelligence tools.

COMPUTERS IN BIOLOGY AND MEDICINE (2022)

Add to Collection

Article Biochemistry & Molecular Biology

MLACP 2.0: An updated machine learning tool for anticancer peptide prediction

Le Thi Phan et al.

Computational and Structural Biotechnology Journal (2022)

Add to Collection

Article Chemistry, Medicinal

ProAll-D: protein allergen detection using long short term memory-a deep learning approach

Pallavi M. Shanthappa et al.

Summary: An allergic reaction is the immune system's overreacting to a previously encountered molecule, often a protein, leading to various symptoms. In this study, a deep learning model LSTM was used to predict allergenicity, achieving an accuracy of 91.5% through training and testing with different machine learning techniques and protein sequence descriptors.

ADMET AND DMPK (2022)

Add to Collection

Review Multidisciplinary Sciences

Biological Sequence Classification: A Review on Data and General Methods

Chunyan Ao et al.

Summary: The rapid growth of biological sequences has driven the application of machine learning in this field, focusing on function and modification classification. Establishing a support website to provide information and datasets for classification methods, discussing current challenges and future prospects.

RESEARCH (2022)

Add to Collection

Article Mathematical & Computational Biology

Risk prediction of diabetes and pre-diabetes based on physical examination data

Yu-Mei Han et al.

Summary: This study collected physical examination data and built classification models to enable early diagnosis of diabetes and identify related risk factors.

MATHEMATICAL BIOSCIENCES AND ENGINEERING (2022)

Add to Collection

Article Biochemical Research Methods

AlgPred 2.0: an improved method for predicting allergenic proteins and mapping of IgE epitopes

Neelam Sharma et al.

Summary: AlgPred 2.0 is a web server developed for predicting allergenic proteins and regions in a protein, using various approaches and techniques for model training and validation, achieving high prediction accuracy.

BRIEFINGS IN BIOINFORMATICS (2021)

Add to Collection

Article Biology

DBCOVP: A database of coronavirus virulent glycoproteins

Susrita Sahoo et al.

Summary: DBCOVP is the first manually curated, web-based resource providing comprehensive information on structural virulent glycoproteins from coronavirus genomes. The database offers various sequence-structural properties for users to browse and analyze information in different ways, as well as predicted T-cell and B-cell epitopes that may play a significant role in immune responses. The database also provides an easy-to-use interface with built-in tools for a variety of analyses, making it an important resource for coronavirus research and vaccine development.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Add to Collection

Article Biology

In-silico identification of subunit vaccine candidates against lung cancer-associated oncogenic viruses

Anjali Lathwal et al.

Summary: This study developed subunit vaccine candidates against lung cancer-causing oncogenic viruses using a reverse vaccinology approach. Through systematic analysis of protein components from nine oncogenic virus species, 125 best antigenic epitopes were identified with predicted B-cell, T-cell, and/or MHC-binding capability, as well as vaccine adjuvant potential. These epitopes show promising immunogenic potential for future development as vaccines against lung cancer-causing viruses.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Risk Prediction of Diabetes: Big data mining with fusion of multifarious physical examination indicators

Hui Yang et al.

Summary: The study designed a computational system to predict diabetes risk by combining various types of physical examination data. Statistical analysis was conducted on different physical examination indexes to develop a model that can distinguish diabetes patients from healthy individuals. A diabetes risk scorecard was established to improve the convenience and flexibility of the model. Lastly, an online diabetes risk assessment system was set up to enhance diabetes cascade screening and personal lifestyle management.

INFORMATION FUSION (2021)

Add to Collection

Article Biochemistry & Molecular Biology

iLearnPlus: a comprehensive and automated machine-learning platform for nucleic acid and protein sequence analysis, prediction and visualization

Zhen Chen et al.

Summary: iLearnPlus is the first machine-learning platform with graphical- and web-based interfaces for analysis and predictions using nucleic acid and protein sequences, providing a comprehensive set of algorithms and automating sequence-based feature extraction and analysis. It caters to experienced bioinformaticians and biologists with no programming background, showcasing its capabilities through case studies on lncRNA prediction and crotonylation site prediction.

NUCLEIC ACIDS RESEARCH (2021)

Add to Collection

Article Biology

ChAlPred: A web server for prediction of allergenicity of chemical compounds

Neelam Sharma et al.

Summary: A method for predicting the allergenic potential of chemical compounds and designing chemical analogs with desired allergenicity was developed and validated using machine learning approaches. The best performing model achieved a maximum accuracy of 83.39% and an AUC of 0.93 on the validation dataset. A web server, ChAlPred, was created to allow researchers to predict and design chemicals with allergenic properties.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Add to Collection

Article Biochemical Research Methods

DeepFusion-RBP: Using Deep Learning to Fuse Multiple Features to Identify RNA-binding Protein Sequences

Xu Wang et al.

Summary: The study introduces a deep learning framework DeepFusion-RBP that cuts RNA sequences with a sliding window method and customizes models for different features, achieving accurate classification of RNA-binding proteins.

CURRENT BIOINFORMATICS (2021)

Add to Collection

Article Biochemical Research Methods

Computational prediction and interpretation of cell-specific replication origin sites from multiple eukaryotes by exploiting stacking framework

Leyi Wei et al.

Summary: The study introduced a novel machine learning approach called Stack-ORI to identify replication origin sites (ORIs) in four different eukaryotic species. Results showed that Stack-ORI outperformed baseline models on both training and independent datasets, consistently achieving better performance across all cell-specific models. The novel approach also provided necessary explanations for model success, highlighting the most important feature encoding schemes significant for predicting cell-specific ORIs.

BRIEFINGS IN BIOINFORMATICS (2021)

Add to Collection

Article Biotechnology & Applied Microbiology