Related Experiment Video
Updated: Oct 24, 2025
10:37
Deep Proteome Profiling by Isobaric Labeling, Extensive Liquid Chromatography, Mass Spectrometry, and Software-assisted Quantification
Published on: November 15, 2017
12.2K
DeepSec: a deep learning framework for secreted protein discovery in human body fluids.
Dan Shao1,2,3, Lan Huang1, Yan Wang1,4
1Key laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education, College of Computer Science and Technology, Jilin University, Changchun 130012, China.
Bioinformatics (Oxford, England)
|August 16, 2021
Summary
We developed DeepSec, a deep learning framework to identify secreted proteins in human body fluids for disease biomarker discovery. This method improves upon existing techniques and identified 104 potential kidney cancer markers.
Area of Science:
- Proteomics
- Bioinformatics
- Machine Learning
Background:
- Secreted human proteins in body fluids are valuable disease indicators.
- Proteomics research faces challenges in comprehensive profiling due to protein complexity and technical limitations, leading to data discrepancies.
- A well-defined proteomics landscape across human fluids is currently lacking.
Purpose of the Study:
- To develop a computational framework for identifying secreted proteins across diverse human body fluids.
- To address the limitations in current proteomics platforms for comprehensive fluid profiling.
- To facilitate biomarker discovery through accurate identification of secreted proteins.
Main Methods:
- Developed DeepSec, a deep learning framework utilizing Convolutional Neural Networks and Bidirectional Gated Recurrent Units for sequence-based protein identification.
- Applied an end-to-end approach for classifying secreted proteins in 12 types of human body fluids.
- Validated performance using ROC curves, achieving an average AUC of 0.85-0.94.
Main Results:
- DeepSec accurately identifies secreted proteins across 12 human body fluid types.
- The framework demonstrates superior performance compared to existing state-of-the-art methods, particularly for blood proteins.
- A case study on kidney cancer identified 104 potential marker proteins using DeepSec and TCGA genomics data.
Conclusions:
- DeepSec offers a robust and accurate method for identifying secreted proteins in human body fluids.
- The framework has significant potential for advancing biomarker discovery in various diseases.
- DeepSec provides a valuable tool for researchers in proteomics and clinical diagnostics.
