Feature Planning

[mmzdouc]

Add

• Documentation: Sphinx-based auto-docks deployed to GitHub pages (library-documentation)
• Quant Phenotype: correlation analysis between features as additional stringency filter (inside spec sim group or independent)
• Group comparison: implement statistical testing of feature abundances between groups
• Read params from session file too
• Write phenotype results into .csv-files (see FERMO output with numeric values FERMO#54)

Change

• Rel int trace pseudochromatogram: round to three post-comma digits to vis low-int features
• Investigate how to reduce the size of the created JSON file (esp. sample-specific features): which data can be omitted? Whitespace/carriage returns? Pre-calculate a trace for general feature only?