Protromics-data-analysis: iPOP Study- Patient Z

This project explores proteomics data from the Integrated Personal Omics Profiling (iPOP) study conducted at Stanford University. The iPOP study is a pioneering effort in longitudinal omics profiling aimed at personalizing precision medicine. Here, we focus specifically on Patient Z, analyzing protein abundance patterns during an infection and recovery phase.

Background

The iPOP study tracks 106 individuals over several years, capturing biological samples at multiple timepoints across states of health and disease. Many participants, including Patient Z, are pre-diabetic, enabling insights into disease progression and immune response. This analysis focuses on proteomics data during a known infection period for Patient Z, helping us understand molecular changes during illness and recovery.

Dataset

File: abundance.csv

• Dimensions: 950 samples × 302 proteins
• Columns: Protein names
• Rows: Timepoints (named or numerical) when samples were taken

Focus Timepoints:

• Infection_Early
• Infection_Middle
• Infection_Late
• Infection_Recovery_Early
• Infection_Recovery_Late

---

Methods

1. Data Loading & Sanity Check

• Loaded proteomics abundance data using pandas.
• Verified most abundant protein in each infection stage sample:
  • Albumin (ALB) was the most abundant protein — biologically expected, as it's the most abundant protein in human blood.

---

2. Normalization: Relative Abundance

• Scaled each protein’s values between 0 and 1:
  • Subtracted the minimum abundance
  • Divided by the protein’s range (max - min)

relative = (data - data.min()) / (data.max() - data.min())

3. Visualization: Heatmap

• Plotted a Seaborn heatmap for the first 20 proteins across the 5 infection-related samples.
• The heatmap visually reveals changes in protein abundance during the infection and recovery timeline.

---

Observations

• The infection recovery phase shows a clear shift in protein abundance profiles.
• CFD (Complement Factor D) exhibited:
  • Low abundance during the infection.
  • Increased abundance during recovery.

Clinical Insight:
CFD deficiency is associated with recurrent bacterial infections — this observed pattern aligns with known immune response dynamics during recovery.

References

Chen, R., et al. (2012). Personal Omics Profiling Reveals Dynamic Molecular and Medical Phenotypes. Cell, 148(6), 1293–1307. https://doi.org/10.1016/j.cell.2012.02.009