v0.3.0

New Rework of the package!

Key Updates:

• Added ggplot Plotting to PCoA
  The function compute_beta_diversity() now incorporates ggplot for visualizing PCoA plots. This update enhances plot aesthetics and provides greater flexibility for further customization of beta diversity visualizations.

• Optimized Function Performance
  All major functions—compute_wilcoxon_and_plot(), compute_beta_diversity(), and compute_volcano_plots()—have been significantly smoothened for optimal processing. This results in faster computations and more efficient memory usage, ensuring scalability even with large datasets.

• New Plot Ordering Option for compute_wilcoxon_and_plot()
  A new parameter has been added to allow users to control the ordering of plots in the generated PDF file. Users can now choose between:

  • "kruskal" (default): Orders the boxplots based on Kruskal-Wallis p-values, displaying the most significant comparisons at the beginning.
  • "rownames": Maintains the original row order of the input data.frame, allowing users to have more control over the output layout.

• Pattern-Filled Boxplots in compute_wilcoxon_and_plot()
  Introduced the option to fill boxplots with patterns (stripes, crosshatch, etc.) using ggpattern. This is especially useful for enhancing visual differentiation in grayscale plots or for publications where color is not ideal.

• Fixed Level Ordering in Barplots
  Corrected improper level ordering in the intra-group variance barplots generated by the compute_beta_diversity() function. The barplot levels now respect the grouping structure for clearer interpretation of variance within groups.

• New rowMedians() Function
  Introduced the rowMedians() function to efficiently calculate the median for each row of a data.frame, providing a robust measure of central tendency for row-wise data summaries.

• Fixed se.median() Function
  Addressed a bug in the se.median() function, which was not functioning properly in previous versions. This function now works as intended for calculating the standard error of the median.

• Reworked compute_volcano_plots() Function
  Major improvements to the compute_volcano_plots() function. The calculation of Log-Fold Change (LFC) has been updated with a new formula that takes into account both the median and the mean of the distribution. Users now have full control over which central tendency to prioritize (mean, median, or both equally balanced). We recommend consulting the function's documentation for further details on this new LFC formula.

• Improved mkdir() Function
  The mkdir() function has been enhanced to support recursive directory creation, allowing users to create nested subfolders in one command, saving time and reducing potential errors.

• Updated Citation and Package Attach Message
  Adjusted the package's citation format to include appropriate metadata. A package startup message has been introduced, reminding users to cite microbAIDeR in their publications and providing a link to the package's documentation.

• New Group Statistics in Excel Output
  The Excel files generated by compute_wilcoxon_and_plot() and compute_volcano_plots() now include both the median and the standard error of the median for each group. Previously, only the mean and standard error of the mean were reported. This change provides more comprehensive statistical summaries for taxa comparisons.

v0.2.2

1. Fixed minor bugs
2. Now adonis exportes all the details, including R2
3. Added mkdir function
4. Implemented the possibility of adding transparency to PCoA points through the alpha.points parameter.

v0.2.1

Minor bug fixes causing problems with the trends evaluation.
Improved the microbaider_install_dependancies() function

v0.2.0

Version v0.2.0 - A whole new level

• Introduced the new function to plot volcano plots ( i.e., compute_volcano_plots() )
• Changed the tabular output of the Wilcoxon P-values, now it's not tidy anymore but much easier to read and interpret
• Added the possibility to provide a manual comparison list for multiple-group analyses with the compute_wilcoxon_and_plot_function()
• Corrected bugs concerning the load of openxlsx package
• Fixed small bugs causing conflicts with other packages
• The old sigFunc() function is now called sigFunction() and now allows to choose whether or not to plot a symbol for trends.
• Added a paired flag in the main functions to specify whether to perform paired signed-rank Wilcoxon tests
• Corrected few typos in the RDocumentation

Here the source package for manual install. Otherwise read the README.md

v0.1.2

Minor bug fixes to RDocumentation.
Added the new function scale.manual

v0.1.1

Minor bug fixes for the major functions
Added new minor features for graph customization

v0.1.0

First release of this R-package