count_matrix.txt

• First row: sample IDs
• First column: feature names
• Row name format (gene-level): <gene_id>|<gene_name>|<gene_type>
• Row name format (domain-level): <gene_id>|<gene_name>|<gene_type>|<domain_id>

feature_importances.txt

• First row: None
• Column 1: feature name
• Column 2: feature importance

metrics.txt

• First row: None
• Column 1: run index
• Column 2: test ROAUC
• Column 3: train ROAUC