Protrec2

Protrec2 integrates tissue-specific protein complex annotations with Bayesian inference to recover unreported but biologically present proteins in proteomics data.

Key Features:

• Probabilistic Framework: Protrec2 employs Bayesian inference combined with tissue-specific protein complex annotations for missing protein recovery.
• Benchmarking Against Established Methods: Protrec2 was evaluated against PROTein RECovery, Functional Class Scoring, Hypergeometric Enrichment, and Gene Set Enrichment Analysis.
• Dual Evaluation Scenarios: Protrec2 was tested under upper-bound and lower-bound scenarios reflecting distinct real-world use cases.
• High Recovery Rates: In upper-bound evaluations, Protrec2 achieved up to 98.4% recovery in A549 and 96.5% in HeLa, validating 650 and 453 proteins respectively.
• Maintained Precision: In lower-bound evaluations, Protrec2 validated over 90% of predicted proteins in A549 and 74.6% in HeLa.
• Clinical Validation: Protrec2 predictions were validated against CPTAC data in lung tumor-normal pairs.

Scientific Applications:

• Discovery Proteomics: Protrec2 enables recovery of biologically present proteins that remain unreported due to technical limitations in proteomics experiments.
• Cancer Proteomics: Protrec2 identifies cancer-specific proteins such as P4HA3, SNX1, HIP1R, and NOS2 that are known to play key roles in lung cancer.
• Tumor-Normal Comparative Analysis: Protrec2 distinguishes cancer-specific proteins that are mostly upregulated from normal-exclusive proteins that are downregulated.
• Translational Research: Protrec2 provides biologically grounded predictions with demonstrated clinical relevance for advancing from discovery to clinical applications.

Methodology:

Protrec2 integrates tissue-specific protein complex annotations with Bayesian inference. The method was benchmarked on HeLa and A549-derived proteomes under upper-bound and lower-bound scenarios, and applied to six matched lung tumor-normal pairs with validation against CPTAC.