Choice of High-Throughput Proteomics Method Affects Data Integration with Transcriptomics and the Potential Use in...

Area of Science:

• Biochemistry
• Genomics
• Proteomics

Background:

• Breast cancer (BC) classification and treatment have advanced, yet overdiagnosis, overtreatment, and recurrence remain significant challenges.
• Accurate BC subtyping and identification of prognostic markers are crucial for effective personalized medicine.

Purpose of the Study:

• To develop and validate a protocol for parallel transcriptome and proteome analysis of breast cancer tissue samples.
• To evaluate different mass spectrometry acquisition and data processing methods for optimal results.
• To explore the potential of integrated multi-omics data for biomarker discovery and BC subtyping.

Main Methods:

• Development of a semi-automated protein digestion protocol integrated with RNA extraction.
• Parallel mass spectrometry-based transcriptome and proteome analysis using Data Dependent Acquisition (DDA) and Data Independent Acquisition (DIA) on 116 BC samples.
• Comparative analysis of data processing tools (MaxQuant, EncyclopeDIA, DIA-NN) and Gene Set Enrichment Analysis (GSEA).

Main Results:

• The DIA-NN software demonstrated superior protein identification, reproducibility, and correlation with RNA-seq data compared to other methods.
• Gene Set Enrichment Analysis revealed complementary insights from transcriptomic and proteomic data.
• A decision tree model utilizing differentially abundant proteins successfully predicted BC intrinsic subtypes and clinical features (ER, HER2 status, proliferation, aggressiveness).

Conclusions:

• The developed parallel transcriptome and proteome analysis protocol is effective for breast cancer tissue samples.
• Integrated multi-omics data analysis provides complementary information for a comprehensive understanding of BC biology.
• The identified protein signatures hold promise as potential biomarkers for BC classification and personalized medicine strategies.