API Reference

Calculator

class py50.Calculator(data: DataFrame, name_col: str = None, concentration_col: str = None, response_col: str | list = None)[source]
__init__(data: DataFrame, name_col: str = None, concentration_col: str = None, response_col: str | list = None)[source]
calculate_absolute_ic50(name_col: str = None, concentration_col: str = None, response_col: str | list = None, input_units: str = None, verbose: bool = None)[source]

Calculations previously performed in absolute_calculation(). The dictionary results are converted into a pandas DataFrame

Parameters:

  • name_col – str Name column from DataFrame

  • concentration_col – str Concentration column from DataFrame

  • response_col – Union[str, list] Response column from DataFrame. Can be a single column (i.e. already a calculated average) or a list of columns to be averaged. The columns will be averaged internally within the function.

  • input_units – str Units of input dataset. Default is nM.

  • verbose – bool Output drug concentration units.

Returns:

DataFrame generated from the list from the absolute_calculation method

calculate_ic50(name_col: str = None, concentration_col: str = None, response_col: str | list = None, input_units: str = None, verbose: bool = None)[source]

Calculations previously performed in relative_calculation(). The dictionary results are converted into into a pandas DataFrame

Parameters:

  • name_col – str Name column from DataFrame

  • concentration_col – str Concentration column from DataFrame

  • response_col – Union[str, list] Response column from DataFrame. Can be a single column (i.e. already a calculated average) or a list of columns to be averaged. The columns will be averaged internally within the function.

  • input_units – str Units of input dataset. Default is nM.

  • verbose – bool Output drug concentration units.

Returns:

DataFrame generated from the list from the relative_calculation method

calculate_pic50(name_col: str = None, concentration_col: str = None, response_col: str | list = None, input_units: str = None, verbose: bool = None)[source]

Convert IC50 into pIC50 values. Calculation is performed using the absolute_calculation. As such, two columns will be appended - relative pIC50 and absolute pIC50. Conversion is performed by convert the IC50 values from nM to M levels and then taking the negative log value of said number.

Parameters:

  • name_col – str Name column from DataFrame

  • concentration_col – str Concentration column from DataFrame

  • response_col – Union[str, list] Response column from DataFrame. Can be a single column (i.e. already a calculated average) or a list of columns to be averaged. The columns will be averaged internally within the function.

  • input_units – str Units of input dataset. Default is nM.

  • verbose – bool Output drug concentration units.

Returns:

DataFrame from calculate_absolute_ic50 along with the pIC50 values

show(rows: int = None)[source]

show DataFrame

Parameters:

rows – int Indicate the number of rows to display. If none, automatically show 5.

Returns:

DataFrame

show_column(key: str = None)[source]

View specific column from DataFrame

Parameters:

key – String Input column header. Must be a column header found in class input DataFrame.

Returns:

DataFrame

to_csv(path: str = None, index: bool = False, **kwargs)[source]

Save the tabel to csv file.

Parameters:

  • path – str Designate the save path for the table.

  • index – bool Write the row names.

  • kwargs – Keyword arguments for pandas.DataFrame.to_csv.

Returns:

csv file

PlotCurve

class py50.PlotCurve(data: DataFrame, name_col: str | None = None, concentration_col: str | None = None, response_col: str | list | None = None)[source]

Bases: object

__init__(data: DataFrame, name_col: str | None = None, concentration_col: str | None = None, response_col: str | list | None = None)[source]
curve_plot(concentration_col: str = None, response_col: str | list = None, name_col: str = None, query: str = None, title: str = None, titlesize: int = 16, xlabel: str = None, ylabel: str = None, axis_fontsize: int = 14, conc_unit: str = 'nM', xscale: str = 'log', xscale_ticks: tuple = None, ymax: int = None, ymin: int = None, line_color: str = 'black', line_width: int = 1.5, errorbar: str = 'sd', marker: bool = None, markersize: int = 8, legend: bool = False, legend_loc: str = 'best', box: bool = False, box_color: str = 'gray', box_intercept: int = 50, conc_target: int = None, hline: int = None, hline_color: str = 'gray', vline: int = None, vline_color: str = 'gray', figsize: tuple = (6.4, 4.8), savepath: str = None, verbose: bool = None, **kwargs)[source]

Generate a dose-response curve for a single drug target. Because a data table can contain multiple drugs, user must specify specific target.

Parameters:

  • concentration_col – str Concentration column from DataFrame.

  • response_col – Union[str, list] Response column from DataFrame.

  • name_col – str Column containing drug name for plotting.

  • query – str Draw a curve for a specific query in the dataset. Only needed when response_col is a list.

  • title – str Title of the figure.

  • titlesize – tuple Modify plot title font size.

  • xlabel – str Title of the X-axis.

  • ylabel – str Title of the Y-axis.

  • axis_fontsize – int Modify axis label font size.

  • conc_unit – str Input unit of concentration. Can accept nanomolar (nM) and micromolar (uM or µM). If the units are different, for example in the DataFrame units are in nM, but the units for the graph are µM, the units from the DataFrame will be converted to match the conc_unit input. The final plot will scale based on the conc_unit input. By default, it will assume input concentration will be in nM.

  • xscale – int Set the scale of the X-axis as logarithmic or linear. It is logarithmic by default.

  • xscale_ticks – tuple Set the scale of the X-axis

  • ymax – int Give a set maximum limit for the Y-Axis

  • ymin – int Give a set minimum limit for the Y-Axis

  • line_color – str. Takes a list of colors. By default, it uses the CBPALETTE. List can contain name of colors or colors in hex code.

  • line_width – int Set width of lines in plot.

  • errorbar – str Set the type of seaborn errorbar to use. Defaults to ‘sd’.

  • marker – Optional, list Takes a list of for point markers.

  • markersize – int Set the marker size.

  • legend – Optional, bool Denotes a figure legend.

  • legend_loc – str Determine legend location. Default is best. Matplotlib options can be found here https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.legend.html

  • box – Optional bool. Draw a box to highlight a specific location. If box = True, then the box_color, box_intercept, and x_concentration MUST ALSO BE GIVEN.

  • box_color – str Set color of box. Default is gray.

  • box_intercept – int Set horizontal location of box. By default, it is set at 50% of the Y-axis.

  • conc_target – int Set vertical location of the box. By default, this is set to None. For example, if the box_intercept is set to 50%, then the x_concentration must be the Absolute IC50 value. If there is an input to x_concentration, it will override the box_intercept and the response data will move accordingly. Finally, the number must be in the same unit as the X-axis. i.e., if the axis is in µM, then the number for the x_concentration should be in µM and vice versa.

  • hline – Int or float Draw a horizontal line across the graph. This line will stretch across the length of the plot. This is optional and set to 0 by default.

  • hline_color – str Set color of horizontal line. Default color is gray.

  • vline – int or float This line will stretch across the height of the plot. This is optional and set to 0 by default.

  • vline_color – str Set color of vertical line. Default color is gray.

  • figsize – tuple Set figure size.

  • savepath – str File path for save location.

  • verbose – bool Output information about the plot.

Returns:

Figure

grid_curve_plot(concentration_col: str = None, response_col: str = None, name_col: str = None, column_num: int = 2, title: str = None, titlesize: int = 20, xlabel: str = None, ylabel: str = None, conc_unit: str = 'nM', xscale: str = 'log', xscale_ticks: tuple = None, ymax: int = None, ymin: int = None, line_color: list = ('#000000', '#E69F00', '#56B4E9', '#009E73', '#F0E442', '#0072B2', '#D55E00', '#CC79A7'), line_width: int = 1.5, box: bool = False, box_color: str = 'gray', box_intercept: int = 50, hline: int = None, hline_color: str = 'gray', vline: int = None, vline_color: str = 'gray', figsize: tuple = (8.4, 4.8), savepath: str = None, verbose: bool = None, **kwargs)[source]

Generate a dose-response curve for mutliple drugs. Each curve will be placed in its own plot which is then placed in a grid.

Parameters:

  • concentration_col – str Concentration column from DataFrame

  • response_col – str Response column from DataFrame

  • name_col – str Name column from DataFrame

  • column_num – int Set number of column grid

  • title – str Title of the figure

  • titlesize – int Modify plot title font size

:param xlabel:str

Title of the X-axis

Parameters:

  • ylabel – str Title of the Y-axis

  • ymax – int Give a set maximum limit for the Y-Axis

  • ymin – int Give a set minimum limit for the Y-Axis

  • conc_unit – str Input will assume that the concentration will be in nM. Thus, it will be automatically converted into µM. If xscale_unit is given as nM, no conversion will be performed.

  • xscale – str Set the scale of the X-axis as logarithmic or linear. It is logarithmic by default.

  • xscale_ticks – tuple Set the scale of the X-axis

  • line_color – list Takes a list of colors. By default, it uses the CBPALETTE. List can contain name of colors or colors in hex code.

  • line_width – int Set width of lines in plot.

  • errorbar – Set the type of seaborn errorbar to use. Defaults to ‘sd’.

  • box – bool Draw a box to highlight a specific location. If box = True, then the box_color, and box_intercept MUST ALSO BE GIVEN.

  • box_color – str Set color of box. Default color is gray.

  • box_intercept – int Set horizontal location of box. By default, it is set at Absolute IC50.

  • hline – int or float Draw horizontal line that will stretch across the length of the plot. This is optional and set to 0 by default.

  • hline_color – str Set color of horizontal line. Default color is gray.

  • vline – int or float Draw a line that will stretch across the height of the plot. This is optional and set to 0 by default.

  • vline_color – str Set color of vertical line. Default color is gray.

  • figsize – tuple Set figure size for subplot.

  • savepath – str File path for save location.

  • verbose – bool Output information about the plot.

Returns:

Figure

multi_curve_plot(concentration_col: str = None, response_col: str = None, name_col: str = None, title: str = None, titlesize: int = 12, xlabel: str = None, ylabel: str = None, conc_unit: str = 'nM', xscale: str = 'log', xscale_ticks: tuple = None, ymax: int = None, ymin: int = None, axis_fontsize: int = 10, line_color: list = ('#000000', '#E69F00', '#56B4E9', '#009E73', '#F0E442', '#0072B2', '#D55E00', '#CC79A7'), marker: list = ('o', '^', 's', 'D', 'v', '<', '>', 'p'), markersize: int = 8, line_width: int = 1.5, errorbar: str = 'sd', legend: bool = False, legend_loc: str = 'best', box_target: str = None, box_color: str = 'gray', box_intercept: int = 50, hline: int = None, hline_color: str = 'gray', vline: int = None, vline_color: str = 'gray', figsize: tuple = (6.4, 4.8), savepath: str = None, verbose: bool = None, **kwargs)[source]

Generate a dose-response plot for multiple drug targets. Curves will be placed into a single plot.

Parameters:

  • concentration_col – str Concentration column from DataFrame

  • response_col – str Response column from DataFrame

  • name_col – Column containing name of drug from DataFrame

  • title – str Title of the figure

  • titlesize – int Modify plot title font size

  • xlabel – str Title of the X-axis

  • ylabel – str Title of the Y-axis

  • conc_unit – str Input will assume that the concentration will be in nM. Thus, it will be automatically converted into µM. If xscale_unit is given as nM, no conversion will be performed.

  • xscale – str Set the scale of the X-axis as logarithmic or linear. It is logarithmic by default.

  • xscale_ticks – tuple Set the scale of the X-axis

  • ymax – int Give a set maximum limit for the Y-Axis

  • ymin – int Give a set minimum limit for the Y-Axis

:param axis_fontsize:int

Modify axis label font size

Parameters:

  • line_color – str Takes a list of colors. By default, it uses the CBPALETTE. List can contain name of colors or colors in hex code.

  • markersize – Set the marker size.

  • line_width – int Set width of lines in plot.

  • errorbar – Set the type of seaborn errorbar to use. Defaults to ‘sd’.

  • marker – list Takes a list for point markers. Marker options can be found here: https://matplotlib.org/stable/api/markers_api.html

  • legend – bool Denotes a figure legend.

  • legend_loc – str Determine legend location. Matplotlib options can be found here https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.legend.html

  • box_target – str Draw a box to highlight a specific drug curve. Must use specific drug name.

  • box_color – str Set color of box. Default color is gray.

  • box_intercept – int Set horizontal location of box. By default, it is set at Absolute IC50.

  • hline – int or float Draw a horizontal line that will stretch across the length of the plot. This is optional and set to 0 by default.

  • hline_color – str Set color of horizontal line. Default color is gray.

  • vline – int or float Draw a line that will stretch across the height of the plot. This is optional and set to 0 by default.

  • vline_color – str Set color of vertical line. Default color is gray.

  • figsize – tuple Set figure size.

  • savepath – str File path for save location.

  • verbose – bool Output information about the plot.

Returns:

Figure

show(rows: int = None)[source]

show DataFrame

Parameters:

rows – int Indicates the number of rows to display. If none, automatically show 5.

Returns:

DataFrame

to_csv(path: str = None, index: bool = False, **kwargs)[source]

Save table to csv file. :param path: str

Designate save path for table.

Parameters:

index – bool Write row names.

**kwargs

Keyword arguments for pandas.DataFrame.to_csv.

Returns:

Stats

class py50.Stats(data)[source]

Class contains wrappers for pingouin module. The functions output data as a Pandas DataFrame. This is in a format needed for plotting with functions in class Plots(), however they can also be used individually to output single DataFrame for output as a csv or xlsx file using pandas.

__init__(data)[source]
static explain_significance()[source]

Print out DataFrame containing explanations for star values. This is used for reference. See [GraphPad](https://www.graphpad.com/support/faq/what-is-the-meaning-of–or–or–in-reports-of-statistical-significance-from-prism-or-instat/)

Returns:

pandas.DataFrame

get_anova(value_col=None, group_col=None, **kwargs)[source]

One-way and N-way ANOVA.

Parameters:
Returns:

Pandas.DataFrame

get_cochran(value_col=None, group_col=None, subgroup_col=None)[source]

Calculate Cochran Q Test. This is used when the dependent variable, or value_col, is binary. For details between groups, posthoc test will be needed.

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • group_col – String Name of column containing the within factor.

  • subgroup_col – String Name of column containing the subject identifier.

Returns:

Pandas.DataFrame

get_friedman(group_col=None, value_col=None, subgroup_col=None, method='chisq')[source]

Calculate Friedman Test. Determines if distributions of two or more paired samples are equal. For details between groups, posthoc test (get_pairwise_tests(parametric=False)) will be needed.

Parameters:

  • value_col – String Name of column containing the dependent variable

  • group_col – String Name of column containing the between-subject factor.

  • subgroup_col – String Name of column containing the subject/rater identifier

  • method – String Statistical test to perform. Must be ‘chisq’ (chi-square test) or ‘f’ (F test). See Pingouin documentation for further details

Returns:

Pandas.DataFrame

get_gameshowell(value_col=None, group_col=None, effsize='hedges')[source]

Pairwise Games-Howell post-hoc test

Parameters:
Returns:

Pandas.DataFrame

get_homoscedasticity(value_col=None, group_col=None, method='levene', **kwargs)[source]

Test for data variance.

Parameters:
Returns:

Pandas.DataFrame

get_kruskal(value_col=None, group_col=None, detailed=False)[source]

Calculate Kruskal-Wallis H-test for independent samples.

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • group_col – String Name of column containing the between factor.

  • detailed – Boolean Output additional details from Kruskal-Wallis H-test.

Returns:

Pandas.DataFrame

get_mannu(value_col=None, group_col=None, subgroup_col=None, alternative='two-sided', **kwargs)[source]

Calculate Mann-Whitney U Test. This is a non-parametric version of the independent T-test.

Parameters:

  • self – pandas.DataFrame Input DataFrame.

  • value_col – String Columns containing values for testing.

  • group_col – String Column containing group name.

  • subgroup_col – String Column containing subgroup name.

  • alternative – String Defines the alternative hypothesis, or tail of the test. Must be one of “two-sided”. Must be one of “two-sided” (default), “greater” or “less”.

  • kwargs – Optional Other options available with [pingouin.mwu()](https://pingouin-stats.org/build/html/generated/pingouin.mwu.html)

Returns:

Pandas.DataFrame

get_mixed_anova(value_col=None, group_col=None, within_subject_col=None, subject_col=None, **kwargs)[source]

Mixed-design ANOVA.

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • group_col – String Name of column containing the between factor.

  • within_subject_col – String Name of column containing the within-subject factor (repeated measurements).

  • subject_col – Name of column containing the between-subject identifier.

  • kwargs – optional Other options available with [pingouin.mixed_anova()](https://pingouin-stats.org/build/html/generated/pingouin.mixed_anova.html)

Returns:

Pandas.DataFrame

get_normality(value_col=None, group_col=None, method='shapiro', **kwargs)[source]

Test data normality of dataset.

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • group_col – String Name of columnName of column containing the grouping variable.

  • method – String Normality test. ‘shapiro’ (default). Additional tests can be found with [pingouin.normality()](https://pingouin-stats.org/build/html/generated/pingouin.normality.html)

  • kwargs – optional Other options available with pingouin.normality()

Returns:

Pandas.DataFrame

static get_p_matrix(data, test=None, group_col1=None, group_col2=None, order=None)[source]

Convert dataframe of statistic results into a matrix. Group columns must be indicated. Group 2 is optional and depends on test used (i.e. pairwise vs Mann-Whitney U). Final DataFrame output can be used with the Plots.p_matrix() function to generate a heatmap of p-values.

Parameters:

  • data – pandas.DataFrame Input DataFrame. Must be of already computed test results.

  • group_col1 – String Name of column containing the group

  • group_col2 – String Name of column containing the second group. This variable is optional.

  • test – String Name of the test used to calculate statistics.

  • order – List or String == “alpha” Reorder the groups for the final table. If input is string “alpha”, the order of the groups will be alphabetized.

Returns:

get_pairwise_mixed(value_col=None, group_col=None, within_subject_col=None, subject_col=None, parametric=True, **kwargs)[source]

Posthoc test for mixed ANOVA.

Parameters:
Returns:

pandas.DataFrame

get_pairwise_rm(value_col=None, group_col=None, within_subject_col=None, subject_col=None, parametric=True, **kwargs)[source]

Posthoc test for repeated measures.

Parameters:
Returns:

pandas.DataFrame

get_pairwise_tests(value_col=None, group_col=None, within_subject_col=None, subject_col=None, parametric=True, **kwargs)[source]

Posthoc test for parametric or nonparametric statistics. By default, the parametric parameter is set as True.

Parameters:
Returns:

pandas.DataFrame

get_rm_anova(value_col=None, within_subject_col=None, subject_col=None, correction='auto', detailed=False, effsize='ng2')[source]

One-way and two-way repeated measures ANOVA.

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • within_subject_col – String Name of column containing the within factor.

  • subject_col – String Name of column containing the subject identifier.

  • correction – String or Boolean If True, also return the Greenhouse-Geisser corrected p-value.

  • detailed – Boolean If True, return full ANOVA table.

  • effsize – String Effect size.

Returns:

Pandas.DataFrame

get_tukey(value_col=None, group_col=None, effsize='hedges')[source]

Pairwise Tukey post-hoc test.

Parameters:
Returns:

Pandas.DataFrame

get_welch_anova(value_col=None, group_col=None)[source]

One-way Welch ANOVA

Parameters:

  • value_col – String Name of column containing the dependent variable.

  • group_col – String Name of column containing the grouping variable.

Returns:

Pandas.DataFrame

get_wilcoxon(value_col=None, group_col=None, subgroup_col=None, alternative='two-sided', **kwargs)[source]

Calculate wilcoxon tests. This is non-parametric version of paired T-test. Data number must be uniform to work.

Parameters:

  • value_col – String Columns containing values for testing.

  • group_col – String Column containing group name.

  • subgroup_col – String Column containing subgroup name.

  • alternative – String Defines the alternative hypothesis, or tail of the test. Must be one of “two-sided”. Must be one of “two-sided” (default), “greater” or “less”.

  • kwargs – Optional Other options available with [pingouin.wilcoxon()](https://pingouin-stats.org/build/html/generated/pingouin.wilcoxon.html)

Returns:

Pandas.DataFrame

show(rows=None)[source]

show DataFrame

Parameters:

rows – Int Indicate the number of rows to display. If none, automatically show 5.

Returns:

DataFrame

Plots

class py50.Plots(data)[source]
__init__(data)[source]
barplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', errorbar='sd', capsize=0.1, return_df=None, **kwargs)[source]

Draw a barplot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • errorbar – String Set confidence interval on plot.

  • capsize – Int Set cap size on plot.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns:

boxenplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', return_df=None, **kwargs)[source]

Draw a boxenplot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns:

boxplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', whis=1.5, return_df=None, **kwargs)[source]

Draw a boxplot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • whis – Int Set length of whiskers on plot.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns:

Fig

ci_plot(data: Optional = None, value_col: str = None, group_col: str = None, alpha: float = 0.05, title: str = 'Tukey HSD Confidence Intervals', xlabel: str = None, ylabel: str = None, linewidth: float = 1.5, figsize: tuple = (8, 6), return_stats: bool = False)[source]

Generate a confidence interval plot. The plot utilizes the Tukey Honest Significant Difference (HSD) test and is a wrapper for statsmodels (https://www.statsmodels.org/dev/index.html). ANOVA will also be calculated and its p-value will be plotted alongside the title. :param data: Optional

Input dataset.

Parameters:

  • value_col – str Name of the column containing the dependent variable.

  • group_col – str Name of the column containing the groups.

  • alpha – float The significance level for the test.

  • title – str Set the title for the figure. Defaults to “Tukey HSD Confidence Intervals”.

  • xlabel – str Set the label for the x-axis. If None is given, defaults to the value_col input.

  • ylabel – str Set the label for the y-axis. If None is given, defaults to the group_col input.

  • linewidth – float Set the width of the lines.

  • figsize – tuple Set the figure size. Defaults to (8,6).

  • return_stats – bool Whether to return the Tukey HSD test.

Returns:

distribution(val_col=None, type='histplot', **kwargs)[source]
Parameters:

  • self – Pandas.Dataframe Input data.

  • val_col – String The name of the column containing the dependent variable.

  • type – String The type of figure drawn. For distribution, only “histplot” or “qqplot” supported

  • kwargs – Optional keyword arguments for seaborn or pg.qqplot.

Returns:

figure

static list_test()[source]

List all tests available for plotting

Returns:

p_matrix(data=None, cmap=None, title=None, titlesize=14, linewidths=0.01, linecolor='gray', **kwargs)[source]

Wrapper function for scikit_posthoc heatmap.

Parameters:

  • data – Pandas.Dataframe Input table must be a matrix calculated using the stats.get_p_matrix(). Optional.

  • cmap – List A list of colors. Can be color names or hex codes.

  • title – String Input title for figure.

  • title_titlesize – Int Set size of figure legend.

  • linewidths – Int Set line width of figure.

  • linecolor – String Set line color. Can be color name or hex code.

  • kwargs – Optional Keyword arguemnts associated with [scikit-posthocs](https://scikit-posthocs.readthedocs.io/en/latest/)

Returns:

Pyplot figure

stripplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', return_df=None, **kwargs)[source]

Draw a stripplot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns:

swarmplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', return_df=None, **kwargs)[source]

Draw a swarm plot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns:

violinplot(test=None, group_col=None, value_col=None, group_order=None, subgroup_col=None, subject_col=None, within_subject_col=None, pairs=None, pvalue_label=None, hide_ns=False, palette=None, orient='v', loc='inside', return_df=None, **kwargs)[source]

Draw a violinplot from the input DataFrame.

Parameters:

  • test – String Name of test for calculations. Names must match the test names from the py50.Stats()

  • group_col – String Name of column containing groups. This should be the between depending on the selected test.

  • value_col – String Name of the column containing the values. This is the dependent variable.

  • group_order – List. Place the groups in a specific order on the plot.

  • subgroup_col – String Name of the column containing the subgroup for the group column. This is associated with the hue parameters in Seaborn.

  • subject_col – String Name of the column containing the subject column.

  • within_subject_col – String Name of the column containing the within subject column.

  • pairs – List A list containing specific pairings for annotation on the plot.

  • pvalue_label – List. A list containing specific pvalue labels. This order must match the length of pairs list.

  • hide_ns – bool Automatically hide groups with no significance from plot.

  • palette – String or List. Color palette used for the plot. Can be given as common color name or in hex code.

  • orient – String Orientation of the plot. Only “v” and “h” are for vertical and horizontal, respectively, is supported

  • loc – String Set location of annotations. Only “inside” or “outside” are supported.

  • return_df – Boolean Returns a DataFrame of calculated results. If pairs used, only return rows with annotated pairs.

Returns: