Related Experiment Video
Updated: May 24, 2025
09:18
Isolation of Extracellular Vesicles from Murine Bronchoalveolar Lavage Fluid Using an Ultrafiltration Centrifugation Technique
Published on: November 9, 2018
9.4K
SecProGNN: Predicting Bronchoalveolar Lavage Fluid Secreted Protein Using Graph Neural Network
IEEE Journal of Biomedical and Health Informatics
|March 5, 2025
Summary
A new deep learning model, SecProGNN, accurately predicts secretory proteins in bronchoalveolar lavage fluid (BALF). This tool aids in identifying potential lung adenocarcinoma biomarkers from BALF proteomic data.
Area of Science:
- Proteomics
- Bioinformatics
- Computational Biology
Background:
- Bronchoalveolar lavage fluid (BALF) contains over 3,000 identified proteins, crucial for studying pulmonary diseases.
- Comprehensive characterization of BALF proteins is hindered by complexity and technological limitations.
Purpose of the Study:
- To introduce SecProGNN, a novel deep learning framework for predicting secretory proteins in BALF.
- To leverage SecProGNN for identifying potential biomarkers of lung adenocarcinoma in BALF.
Main Methods:
- Proteins represented as graph-structured data with amino acids as nodes and interactions as edges.
- Graph neural networks (GNNs) used to extract features from protein graphs.
- Multi-layer perceptron (MLP) module employed for the final prediction of BALF secreted proteins.
Main Results:
- SecProGNN framework successfully predicts secretory proteins in BALF.
- Investigation using SecProGNN identified 16 potential protein biomarkers for lung adenocarcinoma.
- The identified candidates are suggested to be secreted into BALF.
Conclusions:
- SecProGNN offers a powerful computational approach for identifying secreted proteins in BALF.
- The identified lung adenocarcinoma candidates warrant further validation as potential diagnostic or prognostic biomarkers.
- This deep learning framework advances the proteomic analysis of BALF for respiratory disease research.
