make_distance_boxplots.py – Creates boxplots to compare distances between categories

Description:

This script creates boxplots that allow for the comparison between different categories found within the mapping file. The boxplots that are created compare distances within all samples of a field value, as well as between different field values. Individual within and between distances are also plotted.

The script also performs two-sample t-tests for all pairs of boxplots to help determine which boxplots (distributions) are significantly different.

Tip: the script tries its best to fit everything into the plot, but there are cases where plot elements may get cut off (e.g. if axis labels are extremely long), or things may appear squashed, cluttered, or too small (e.g. if there are many boxplots in one plot). Increasing the width and/or height of the plot (using –width and –height) usually fixes these problems.

For more information and examples pertaining to this script, please refer to the accompanying tutorial, which can be found at http://qiime.org/tutorials/creating_distance_comparison_plots.html.

Usage: make_distance_boxplots.py [options]

Input Arguments:

Note

[REQUIRED]

-m, --mapping_fp

The mapping filepath

-o, --output_dir

Path to the output directory

-d, --distance_matrix_fp

Input distance matrix filepath (i.e. the result of beta_diversity.py). WARNING: Only symmetric, hollow distance matrices may be used as input. Asymmetric distance matrices, such as those obtained by the UniFrac Gain metric (i.e. beta_diversity.py -m unifrac_g), should not be used as input

-f, --fields

Comma-separated list of fields to compare, where the list of fields should be in quotes (e.g. “Field1,Field2,Field3”)

[OPTIONAL]

-g, --imagetype

Type of image to produce (i.e. png, svg, pdf) [default: pdf]

--save_raw_data

Store raw data used to create boxplots in tab-delimited files [default: False]

--suppress_all_within

Suppress plotting of “all within” boxplot [default: False]

--suppress_all_between

Suppress plotting of “all between” boxplot [default: False]

--suppress_individual_within

Suppress plotting of individual “within” boxplot(s) [default: False]

--suppress_individual_between

Suppress plotting of individual “between” boxplot(s) [default: False]

--suppress_significance_tests

Suppress performing signifance tests between each pair of boxplots [default: False]

-n, --num_permutations

The number of Monte Carlo permutations to perform when calculating the nonparametric p-value in the significance tests. Must be an integer greater than or equal to zero. If zero, the nonparametric p-value will not be calculated and will instead be reported as “N/A”. This option has no effect if –suppress_significance_tests is supplied [default: 0]

-t, --tail_type

The type of tail test to compute when calculating the p-values in the significance tests. “high” specifies a one-tailed test for values greater than the observed t statistic, while “low” specifies a one-tailed test for values less than the observed t statistic. “two-sided” specifies a two-tailed test for values greater in magnitude than the observed t statistic. This option has no effect if –suppress_significance_tests is supplied. Valid choices: low or high or two-sided [default: two-sided]

--y_min

The minimum y-axis value in the resulting plot. If “auto”, it is automatically calculated [default: 0]

--y_max

The maximum y-axis value in the resulting plot. If “auto”, it is automatically calculated [default: 1]

--width

Width of the output image in inches. If not provided, a “best guess” width will be used [default: auto]

--height

Height of the output image in inches [default: 6]

--transparent

Make output images transparent (useful for overlaying an image on top of a colored background) [default: False]

--whisker_length

Length of the whiskers as a function of the IQR. For example, if 1.5, the whiskers extend to 1.5 * IQR. Anything outside of that range is seen as an outlier [default: 1.5]

--box_width

Width of each box in plot units [default: 0.5]

--box_color

The color of the boxes. Can be any valid matplotlib color string, such as “black”, “magenta”, “blue”, etc. See http://matplotlib.sourceforge.net/api/colors_api.html for more examples of valid color strings that may be used [default: same as plot background, which is white unless –transparent is enabled]

--sort

Sort boxplots by increasing median. If no sorting is applied, boxplots will be grouped logically as follows: all within, all between, individual within, and individual between [default: False]

Output:

Images of the plots are written to the specified output directory (one image per field). The raw data used in the plots and the results of significance tests can optionally be written into tab-delimited files (one file per field) that are most easily viewed in a spreadsheet program such as Microsoft Excel.

Compare distances between Fast and Control samples:

This example will generate an image with boxplots for all within and all between distances for the field Treatment, and will also include plots for individual within (e.g. Control vs. Control, Fast vs. Fast) and individual between (e.g. Control vs. Fast). The generated plot PDF and signifiance testing results will be written to the output directory ‘out1’.

make_distance_boxplots.py -d unweighted_unifrac_dm.txt -m Fasting_Map.txt -f "Treatment" -o out1

Only plot individual field value distances:

This example will generate a PNG of all individual field value distances (within and between) for the Treatment field.

make_distance_boxplots.py -d unweighted_unifrac_dm.txt -m Fasting_Map.txt -f "Treatment" -o out2 -g png --suppress_all_within --suppress_all_between

Save raw data:

This example will generate an SVG image of the boxplots and also output the plotting data to a tab-delimited file.

make_distance_boxplots.py -d unweighted_unifrac_dm.txt -m Fasting_Map.txt -f "Treatment" -o out3 -g svg --save_raw_data

Suppress significance tests:

This example will only generate a plot and skip the significance testing step. This can be useful if you are operating on a large dataset and are not interested in performing the statistical tests (or at least not initially).

make_distance_boxplots.py -d unweighted_unifrac_dm.txt -m Fasting_Map.txt -f "Treatment" -o out4 --suppress_significance_tests