Process

padua.process.apply_experimental_design(df, f, prefix='Intensity ')

Load the experimental design template from MaxQuant and use it to apply the label names to the data columns.

Parameters:

• df –
• f – File path for the experimental design template
• prefix –

Returns:

dt

padua.process.build_index_from_design(df, design, remove=None, types=None, axis=1, auto_convert_numeric=True, use_unmatched_index=True)

Build a MultiIndex from a design table.

Supply with a table with column headings for the new multiindex and a index containing the labels to search for in the data.

Parameters:

• df –
• design –
• remove –
• types –
• axis –
• auto_convert_numeric –

Returns:

padua.process.build_index_from_labels(df, indices, remove=None, types=None, axis=1)

Build a MultiIndex from a list of labels and matching regex

Supply with a dictionary of Hierarchy levels and matching regex to extract this level from the sample label

Parameters:

• df –
• indices – Tuples of indices (‘label’,’regex’) matches
• strip – Strip these strings from labels before matching (e.g. headers)
• axis=1 – Axis (1 = columns, 0 = rows)

Returns:

padua.process.combine_expression_columns(df, columns_to_combine, remove_combined=True)

Combine expression columns, calculating the mean for 2 columns

Parameters:

• df – Pandas dataframe
• columns_to_combine – A list of tuples containing the column names to combine

Returns:

padua.process.expand_side_table(df)

Perform equivalent of ‘expand side table’ in Perseus by folding Multiplicity columns down onto duplicate rows

The id is remapped to UID___Multiplicity, which is different to Perseus behaviour, but prevents accidental of non-matching rows from occurring later in analysis.

Parameters:df – Returns:

padua.process.fold_columns_to_rows(df, levels_from=2)

Take a levels from the columns and fold down into the row index. This destroys the existing index; existing rows will appear as columns under the new column index

Parameters:

• df –
• levels_from – The level (inclusive) from which column index will be folded

Returns:

padua.process.get_unique_indices(df, axis=1)

Parameters:

• df –
• axis –

Returns:

padua.process.numeric(s)

Parameters:s – Returns:

padua.process.strip_index_labels(df, strip, axis=1)

Parameters:

• df –
• strip –
• axis –

Returns:

padua.process.transform_expression_columns(df, fn=<Mock id='139759960052512'>, prefix='Intensity ')

Apply transformation to expression columns.

Default is log2 transform to expression columns beginning with Intensity

Parameters:

• df –
• prefix – The column prefix for expression columns

Returns: