Motivation: Accurate protein function prediction is crucial for understanding biological processes and advancing biomedical research. However, the rapid growth of protein sequences far outpaces the experimental characterization of their functions, necessitating the development of automated computational methods.
Results: We present InterLabelGO+, a hybrid approach that integrates a deep learning-based method with an alignment-based method for improved protein function prediction. InterLabelGO+ incorporates a novel loss function that addresses label dependency and imbalance and further enhances performance through dynamic weighting of the alignment-based component. A preliminary version of InterLabelGO+ achieved a strong performance in the CAFA5 challenge, ranking sixth out of 1625 participating teams. Comprehensive evaluations on large-scale protein function prediction tasks demonstrate InterLabelGO+'s ability to accurately predict Gene Ontology terms across various functional categories and evaluation metrics.
Availability And Implementation: The source code and datasets for InterLabelGO+ are freely available on GitHub at https://github.com/QuanEvans/InterLabelGO. A web-server is available at https://seq2fun.dcmb.med.umich.edu/InterLabelGO/. The software is implemented in Python and PyTorch, and is supported on Linux and macOS.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11568131 | PMC |
| http://dx.doi.org/10.1093/bioinformatics/btae655 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
BaNDyT: Bayesian Network Modeling of Molecular Dynamics Trajectories.
J Chem Inf Model
January 2025
Department of Computational and Quantitative Medicine, Beckman Research Institute of the City of Hope, 1218 S 5th Ave, Monrovia, California 91016, United States.
Bayesian network modeling (BN modeling, or BNM) is an interpretable machine learning method for constructing probabilistic graphical models from the data. In recent years, it has been extensively applied to diverse types of biomedical data sets. Concurrently, our ability to perform long-time scale molecular dynamics (MD) simulations on proteins and other materials has increased exponentially.
View Article and Find Full Text PDFStructural and Functional Characterization of the Aorta in Hypertrophic Obstructive Cardiomyopathy.
Circ Heart Fail
January 2025
Aswan Heart Center, Magdi Yacoub Heart Foundation, Egypt (A.M.I., M.R., A. Elsawy, M.H., S.H., W.E., A. Elaithy, A. Elguindy, A. Afifi, Y.A., M.Y.).
Background: Changes in the phenotype and genotype in hypertrophic cardiomyopathy (HCM) are thought to involve the myocardium as well as extracardiac tissues. Here, we describe the structural and functional changes in the ascending aorta of obstructive patients with HCM.
Methods: Changes in the aortic wall were studied in a cohort of 101 consecutive patients with HCM undergoing myectomy and 9 normal controls.
Placental Hypoxia-Induced Ferroptosis Drives Vascular Damage in Preeclampsia.
Circ Res
January 2025
Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada (C.P., S.A., J.W.A., R.L., F.N., J.S., I.C.).
Background: Iron is an essential micronutrient for cell survival and growth; however, excess of this metal drives ferroptosis. Although maternal iron imbalance and placental hypoxia are independent contributors to the pathogenesis of preeclampsia, a hypertensive disorder of pregnancy, the mechanisms by which their interaction impinge on maternal and placental health remain elusive.
Methods: We used placentae from normotensive and preeclampsia pregnancy cohorts, human H9 embryonic stem cells differentiated into cytotrophoblast-like cells, and placenta-specific preeclamptic mice.
Classification and characteristics of bacterial glycosaminoglycan lyases, and their therapeutic and experimental applications.
J Cell Sci
January 2025
National Glycoengineering Research Center, Shandong Key Laboratory of Carbohydrate Chemistry and Glycobiology and State Key Laboratory of Microbial Technology, Shandong University, 72 Binhai Rd, Qingdao, 266237, People's Republic of China.
Glycosaminoglycans (GAGs), as animal polysaccharides, are linked to proteins to form various types of proteoglycans. Bacterial GAG lyases are not only essential enzymes that spoilage bacteria use for the degradation of GAGs, but also valuable tools for investigating the biological function and potential therapeutic applications of GAGs. The ongoing discovery and characterization of novel GAG lyases has identified an increasing number of lyases suitable for functional studies and other applications involving GAGs, which include oligosaccharide sequencing, detection and removal of specific glycan chains, clinical drug development and the design of novel biomaterials and sensors, some of which have not yet been comprehensively summarized.
View Article and Find Full Text PDFDissecting AlphaFold2's capabilities with limited sequence information.
Bioinform Adv
November 2024
Institute of Biochemistry and Molecular Medicine, University of Bern, Bern 3012, Switzerland.
Summary: Protein structure prediction aims to infer a protein's three-dimensional (3D) structure from its amino acid sequence. Protein structure is pivotal for elucidating protein functions, interactions, and driving biotechnological innovation. The deep learning model AlphaFold2, has revolutionized this field by leveraging phylogenetic information from multiple sequence alignments (MSAs) to achieve remarkable accuracy in protein structure prediction.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!