Longitudinal Protein Stability and Individual Specificity Analysis Project

Project Structure

Core Analysis Scripts

• 001-PHI-different-visit.R: Calculate Pearson correlation coefficients for proteins across 6 visit pairs (02-07, 02-12, 02-22, 07-12, 07-22, 12-22)
• 002-PHI-visualization.R: Visualize PHI correlation results with distribution plots and scatter plots
• 003-PHI-grouped-correlation.R: Calculate PHI correlations grouped by sex and age
• 004-PHI-grouped-visualization.R: Visualize grouped PHI results with violin plots and statistical tests
• 005-PHI-pQTL-visualization.R: Analyze relationships between PHI and pQTL effect sizes
• 006-GeneticVariance-PHI-visualization.R: Analyze relationships between PHI and genetic variance
• 007-GxE-PHI-visualization.R: Analyze relationships between PHI and GxE interaction effects
• 008-PHI-same-gene-different-seqid-scatter.R: Compare PHI values for different protein isoforms of the same gene
• 009-protein-class-PHI-visualization.R: Analyze PHI distributions across protein classes (intracellular, extracellular/membrane, secreted)
• 010-protein-sankey-retention.R: Generate Sankey plots showing protein level transitions across visits with retention ratios
• 011-PHI-R2-analysis.R: Analyze relationships between PHI and longitudinal prediction accuracy (R²)

Visit Pairs

The project analyzes protein stability across 4 time points:

• Dataset 1: 2002 (02)
• Dataset 2: 2007 (07)
• Dataset 3: 2012 (12)
• Dataset 4: 2022 (22)

Six visit pairs are analyzed: 02-07, 02-12, 02-22, 07-12, 07-22, 12-22

PHI Bins

Proteins are categorized into PHI bins for analysis:

• 0-0.25: Low stability
• 0.25-0.5: Moderate-low stability
• 0.5-0.75: Moderate-high stability
• 0.75-1: High stability

Requirements

R Packages

• dplyr
• ggplot2
• tidyr
• ggrepel
• cowplot
• ggsignif
• ggalluvial
• lme4
• lmerTest
• parallel

Usage

1. Data Preparation: Ensure input data files are in the correct format and paths are updated in each script
2. Run Analysis Scripts: Execute scripts in numerical order for complete analysis
3. Review Results: Check output directories for generated figures and CSV files

Output

Each script generates:

• CSV files with analysis results
• PNG figures for visualizations
• Summary statistics and test results

Output directories are organized by analysis type:

• results/: Main results directory
• age-sex-seperate-result/: Grouped analysis results
• pQTL-PHI/: pQTL-related analyses
• GeneticVariance-PHI/: Genetic variance analyses
• GxE-PHI/: GxE interaction analyses
• Person-Setpoint/: Prediction accuracy analyses
• Sankey-Plot/: Sankey diagram visualizations

Notes

• File paths in scripts are hardcoded and may need to be adjusted for different systems
• Some scripts require specific input files from previous analysis steps
• Batch effect removal and normalization should be performed before PHI calculations

Citation

If you use this code, please cite the relevant publications and acknowledge the data sources.