v1.1.0

• Compatibility with Python 3.10. Upgrade of statsmodels requirement to version==1.14
• class ScatterTrendlines: input counts dataframe; give the two species to plot against each other to plot_one_graph and the number of trendlines to generate (#104)
• 3 new functions in dict_operations: flatten_dict, super_pop and filter_dict (#104)(#92)
• reorganization Chi2 and other statistical tests (#92)
• SeriesBinning: to put series into bins (#92)

v1.0.0: Upgrade to python version 3.8-3.9

• upgrade to python >= 3.8 EDIT: work with python 3.8 and 3.9 but not 3.10
  Version requirements have been updated for numpy, scikit-bio, scikit-learn, pandas, statsmodels, hdmedians, scipy and plotly (#99)
• mode argument for plot_most_abundant_taxa() and plot_most_prevalent_taxa() that allows to plot the data as boxplot or violin plot, as well as a bargraph. (#89)
• parsers:
  • More formats accepted in parsers: ".xls", ".xlsx", ".odt", ".ods", ".odf", ".xlsb" (#96)
  • Metaphlan3Parser: keep_NCBI_tax_col argument and possibility to compute dataframe with less taxonomical level than expected (#99)
• adapt_phylogenetic_tree_to_counts_df function in utils/phylogenetic_tree_editing.py (#98)
• #93
• generate_report_data for diversity analyses (#88)

v0.7.0

• #78
• Phylogenetic Diversity #81:
  • new argument force_computation allows to force diversity computation even when species are missing in the Tree
  • only accepting trees in skbio.TreeNode
• PlotTaxonomyCounts (#83):
  • new arguments: color_df and sep_series/sep_how added to plot_sample_composition_most_abundant_taxa
  • new plot_most_abundant_taxa method and renaming of the old one to plot_most_prevalant_taxa
  • mode argument for plot_most_abundant_taxa() and plot_most_prevalent_taxa() to plot the data as a boxplot or a violin plot, as well as a bargraph (#89)
• plot_complex_graph in the MatrixBarGraph class to display some metadata below the main graph (#83)
• 2 new arguments, group_col2 and groups2, for the analyse_groups method from the DiversityBase class and its descendants (#86)
• orientation argument for the plot_one_graph method of the GroupBaseGraph class and its descendants (#86)

v0.6.0

• #66
• #69
• Build in visualization for taxonomy counts #72
• Faith's PD and UniFrac #75

v0.5.0

• Visualize metadata cat (#62)
• parser for metaphlan3 (#60)
• pvalue correction in analyse groups (#57)
• add HeatmapGraph (#56)
• Statistical tests for group comparison (#53)
• Mann - Whitney u test in alpha diversity (#50)
• Feature box plot alpha diversity (#46)
• More filtering methods : classes NaNPercentageFiltering and NumberOfDifferentValuesFiltering (#44)

October release

• new features for AlphaDiversity class (#42):

  • visualize method now has two mode: histogram or violin
  • visualize_groups method allows visualization with violin plot based on metadata grouped by a chosen column

• There is now definition of index_col for MetadataParser to define which column to use as the index for the metadata (#42)

• class GenesToTaxonomyto go from gene counts to taxonomy df (#36)

• add TaxonomyRandomSelection to allow random selection on multiindexed dataframe (#32)

• Class AlphaDiversity and child classes ShannonIndex and SimpsonInverseIndex (#29)

• add RandomSelection to normalization module (#28)

Refactoring plot module

• Refactor use of plotting_options : iteration through dict.keys() + exported in BaseGraph (Parent Class)
• plotting_options, output_file and show dealt with in plot_one_graph (instead of at instantiation)
• Refactoring of BarGraph
• Relocation of functions : check_list_of and add_x_to_plotting_options in utils/plot.py
• Relocation of method to bin series in utils/pandas/series.py
• Transition to autosummary in available classes
• Listing plot classes in available classes

Plot module and Filtering refactoring

• Add base for each module (with visualize() method and data_report property)

• Use the new base for MeanFiltering module.

• Refactoring of Filtering class into several classes:

  • NoCountsFiltering that filters on rows or columns with no counts at all

  • NamesFiltering that filters on a given list of row or columns names

    • Either keep the names
    • Or exclude them

  • TaxonomyNamesFiltering that filters on a given list of index names at a chosen level

    • Either keep the names
    • Or exclude them

• Add base module for plots

• Add class BaseGraph and child classes Histogram and BarGraph

v0.1.0 First release

First release of the current work on the package.