Available classes¶

Here are listed the different classes available by module.

Note

We recommend to check the API documentation for more detailed documentation about each class

Parsers¶

How it works?¶

from moonstone.parsers import YourFavouriteParser

parser = YourFavouriteParser("/path/to/the/file")
df = parser.dataframe

List¶

Classic and simple parsers:

MetadataParser

Parse metadata file and allows to apply transformations on them (cleaning…).

Counts parsers (from moonstone.parsers.counts):

`genes.GeneCountsParser`	Common way of representing gene counts per sample in a matrix.
`picrust2.Picrust2PathwaysParser`	Predicted sample pathway abundances output file from Picrust2.

Taxonomy counts parsers (from moonstone.parsers.counts.taxonomy):

`kraken2.SunbeamKraken2Parser`	Parse output from Kraken2 merge table from Sunbeam pipeline.
`metaphlan.Metaphlan2Parser`	Parse output from Metaphlan2 merged table.
`metaphlan.Metaphlan3Parser`	Parse output from Metaphlan3 merged table.
`qiime.Qiime2Parser`	Parse output csv data obtained by Qiime2.

Note

More details on API documentation for api_parsers.

Filtering¶

How it works?¶

from moonstone.filtering import YourFavouriteFiltering

filtering_instance = YourFavouriteFiltering(your_df)
df = filtering_instance.filtered_df

List¶

`basics_filtering.NoCountsFiltering`	Remove rows (default) or columns with no counts.
`basics_filtering.NamesFiltering`	Filtering based on row (default) or column names.
`basics_filtering.NaNPercentageFiltering`	Remove rows (default) or columns with a percentage of NaN values above a given percentage.
`basics_filtering.NumberOfDifferentValuesFiltering`	Filtering of rows (default) or columns based on the number of different (unique) values they hold.
`taxonomy_filtering.TaxonomyNamesFiltering`	Filtering a Taxonomy multiindexed dataframe on index names at a chosen level.
`taxonomy_filtering.TaxonomyMeanFiltering`	Filtering a Taxonomy multiindexed dataframe on sample mean at a chosen level.
`mean_filtering.MeanFiltering`	Remove items with a mean read count below a given threshold.

Note

More details on API documentation for Filtering.

Normalization¶

How it works?¶

from moonstone.normalization import YourFavouriteNormalization

normalization = YourFavouriteNormalization(your_df)
df = normalization.normalized_df

List¶

`counts.geometric_mean.GeometricMeanNormalization`	normalization based on the one performed by DeSeq2.
`counts.total_counts.TotalCountsNormalization`	normalization based on total counts.
`counts.random_selection.RandomSelection`	Randomly select a given number of counts (threshold) among all different items (genes, taxonomical annotation…) for each sample.
`counts.random_selection.TaxonomyRandomSelection`	Allow random selection for taxonomy multi-indexed dataframes.
`processed.scaling_normalization`

Plot¶

How it works?¶

from moonstone.plot import PlotCountsStats

# instantiation
plot_instance = PlotCountsStats(df)

# call one (or more) plotting method(s)
plot_instance.your_favorite_plot()
plot_instance.another_of_your_favorite_plot()

List¶

`counts.PlotCountsStats`	Several plots available to visualize simple count data.
`counts.PlotTaxonomyCounts`	Plots available for taxonomy counts (multiindexed dataframe).
`metadata.PlotMetadataStats`	Several plots available to visualize metadata.

Note

More details on API documentation for Plot.

Analysis¶

How it works?¶

from moonstone.analysis import DifferentialAnalysis

# instantiation
analysis_instance = DifferentialAnalysis(df, metadata_df)
analysis_instance.differential_analysis_by_feature(features, type_of_features, test_to_use, correction_method_used)

Note

The way analysis instances work will change using a defined common way of performing analysis.

List¶

differential_analysis.DifferentialAnalysis

Perform differential analysis using chosen method.

Note

More details on API documentation for Plot.