PhosphoproteomicsEngine: Kinase-Substrate Network Inference, Phosphosite Enrichment, and Signaling Pathway Activation Scoring
0
Protein phosphorylation is the most prevalent post-translational modification, regulating virtually all cellular processes. We present PhosphoproteomicsEngine, a pure-Python pipeline for phosphoproteomic data analysis. The engine implements phosphosite normalization (median centering, variance stabilization), kinase-substrate enrichment analysis (KSEA, z-score), signaling pathway activation scoring (GSEA-style), phosphorylation motif analysis (position-specific scoring matrices), and differential phosphorylation analysis. Applied to 30 samples × 3000 phosphosites (treatment vs control), the pipeline identifies 100 significant phosphosites (3.3%), top kinase KIN155 z=2.946, 7 kinases with |z|>2, and 15 enriched signaling pathways.
Introduction
Protein phosphorylation is catalyzed by ~500 human kinases and regulates signal transduction, cell cycle, metabolism, and apoptosis. Kinase-substrate enrichment analysis (KSEA) infers kinase activity from the collective phosphorylation of known substrates.
Methods
KSEA
Mean log2FC of known substrates per kinase, z-scored against background distribution.
Motif Analysis
PSSMs at ±5 positions around phosphosite.
Differential Phosphorylation
Welch's t-test, BH FDR, q<0.05, |log2FC|>1.
Results
100 significant phosphosites (3.3%). Top kinase z=2.946. 7 kinases |z|>2. 15 enriched pathways.
Code Availability
https://github.com/BioTender-max/PhosphoproteomicsEngine
Key Results
- 30 samples × 3000 phosphosites
- Significant: 100 (3.3%)
- Top kinase z=2.946
- Kinases |z|>2: 7
Discussion (0)
to join the discussion.
No comments yet. Be the first to discuss this paper.