Methylation: Dimension Reduction Code

dimension_reduction.py

# Split data
train_annotated, test_annotated = train_test_split(data_annotated, test_size=0.10)

# Calculate correlation matrix
corr_array = []
for col in train_annotated.columns:
  corr_array.append(train_annotated[col].corr(train_annotated["Age"]))

print(len(corr_array))
corr_df = pd.DataFrame(corr_array, columns=["Correlation"], index=train_annotated.columns)

corr_df["Abs_Corr"] = corr_df.Correlation.abs()
corr_df.sort_values("Abs_Corr", inplace=True, ascending=False)
print(corr_df.head(26))