python-for-biology-week3

Week 3: Hormone and Enzyme Correlation Analysis

Aim

To analyze the relationships between hormone and enzyme levels (Testosterone, LH, ACP, LDH, FSH) in adult male volunteers, and identify significant correlations among these markers.

---

Steps Followed

1. Data Import and Cleaning

   • Loaded the cleaned dataset from Week 2.

2. Exploratory Data Analysis (EDA)

   • Plotted histograms to visualize the distribution of each marker.
   • Created boxplots to identify potential outliers.

3. Descriptive Statistics

   • Generated a summary table including mean, median, standard deviation, min, max, quartiles, and skewness.
   • Noted variables that were skewed to justify the use of Spearman correlation.

4. Correlation Analysis

   • Calculated Spearman correlation between Testosterone and LH:
     • rho = 0.963, p < 0.001 (very strong positive correlation)
   • Calculated Spearman correlation between LDH and ACP:
     • rho = 0.849, p < 0.001 (strong positive correlation)
   • Plotted scatterplots with trend lines for both correlations to visualize relationships.

---

Key Findings

• Testosterone vs LH: Very strong positive correlation, indicating that higher Testosterone levels are associated with higher LH levels in participants.
• LDH vs ACP: Strong positive correlation, indicating that higher LDH activity is associated with higher ACP activity.
• Overall Trends: The Spearman correlation matrix highlighted other relationships among hormones and enzymes, providing insights into their interdependence.
• Skewness in some variables justified the use of Spearman correlation, which is robust to non-normal distributions and outliers.

---

Notes

• All analysis was done in Python using libraries: pandas, numpy, matplotlib, seaborn, and scipy.
• This notebook demonstrates how to perform correlation analysis on biological datasets while maintaining a professional workflow.