Fork me on GitHub

mantel: Apply the Mantel test to two distance matrices

Citations

  • Nathan Mantel. The detection of disease clustering and a generalized regression approach. Cancer research, 27(2 Part 1):209–220, 1967.

  • Karl Pearson. Note on regression and inheritance in the case of two parents. Proceedings of the Royal Society of London, 58:240–242, 1895. doi:https://doi.org/10.1098/rspl.1895.0041.

  • Charles Spearman. The proof and measurement of association between two things. The American journal of psychology, 15(1):72–101, 1904. doi:https://doi.org/10.2307/1422689.

Docstring:

Usage: qiime diversity mantel [OPTIONS]

  Apply a two-sided Mantel test to identify correlation between two distance
  matrices.

  Note: the directionality of the comparison has no bearing on the results.
  Thus, comparing distance matrix X to distance matrix Y is equivalent to
  comparing Y to X.

  Note: the order of samples within the two distance matrices does not need to
  be the same; the distance matrices will be reordered before applying the
  Mantel test.

  See the scikit-bio docs for more details about the Mantel test:

  http://scikit-bio.org/docs/latest/generated/skbio.stats.distance.mantel

Inputs:
  --i-dm1 ARTIFACT     Matrix of distances between pairs of samples.
    DistanceMatrix                                                  [required]
  --i-dm2 ARTIFACT     Matrix of distances between pairs of samples.
    DistanceMatrix                                                  [required]
Parameters:
  --p-method TEXT Choices('spearman', 'pearson')
                       The correlation test to be applied in the Mantel test.
                                                         [default: 'spearman']
  --p-permutations INTEGER
    Range(0, None)     The number of permutations to be run when computing
                       p-values. Supplying a value of zero will disable
                       permutation testing and p-values will not be calculated
                       (this results in *much* quicker execution time if
                       p-values are not desired).               [default: 999]
  --p-intersect-ids / --p-no-intersect-ids
                       If supplied, IDs that are not found in both distance
                       matrices will be discarded before applying the Mantel
                       test. Default behavior is to error on any mismatched
                       IDs.                                   [default: False]
  --p-label1 TEXT      Label for `dm1` in the output visualization.
                                                [default: 'Distance Matrix 1']
  --p-label2 TEXT      Label for `dm2` in the output visualization.
                                                [default: 'Distance Matrix 2']
Outputs:
  --o-visualization VISUALIZATION
                                                                    [required]
Miscellaneous:
  --output-dir PATH    Output unspecified results to a directory
  --verbose / --quiet  Display verbose output to stdout and/or stderr during
                       execution of this action. Or silence output if
                       execution is successful (silence is golden).
  --example-data PATH  Write example data and exit.
  --citations          Show citations and exit.
  --help               Show this message and exit.

Import:

from qiime2.plugins.diversity.visualizers import mantel

Docstring:

Apply the Mantel test to two distance matrices

Apply a two-sided Mantel test to identify correlation between two distance
matrices.  Note: the directionality of the comparison has no bearing on the
results. Thus, comparing distance matrix X to distance matrix Y is
equivalent to comparing Y to X.  Note: the order of samples within the two
distance matrices does not need to be the same; the distance matrices will
be reordered before applying the Mantel test.  See the scikit-bio docs for
more details about the Mantel test:  http://scikit-
bio.org/docs/latest/generated/skbio.stats.distance.mantel

Parameters
----------
dm1 : DistanceMatrix
    Matrix of distances between pairs of samples.
dm2 : DistanceMatrix
    Matrix of distances between pairs of samples.
method : Str % Choices('spearman', 'pearson'), optional
    The correlation test to be applied in the Mantel test.
permutations : Int % Range(0, None), optional
    The number of permutations to be run when computing p-values. Supplying
    a value of zero will disable permutation testing and p-values will not
    be calculated (this results in *much* quicker execution time if
    p-values are not desired).
intersect_ids : Bool, optional
    If supplied, IDs that are not found in both distance matrices will be
    discarded before applying the Mantel test. Default behavior is to error
    on any mismatched IDs.
label1 : Str, optional
    Label for `dm1` in the output visualization.
label2 : Str, optional
    Label for `dm2` in the output visualization.

Returns
-------
visualization : Visualization