Fork me on GitHub

umap: Uniform Manifold Approximation and ProjectionΒΆ

Docstring:

Usage: qiime diversity umap [OPTIONS]

  Apply Uniform Manifold Approximation and Projection.

Inputs:
  --i-distance-matrix ARTIFACT
    DistanceMatrix     The distance matrix on which UMAP should be computed.
                                                                    [required]
Parameters:
  --p-number-of-dimensions INTEGER
    Range(2, None)     Dimensions to reduce the distance matrix to.
                                                                  [default: 2]
  --p-n-neighbors INTEGER
    Range(1, None)     Provide the balance between local and global
                       structure. Low values prioritize the preservation of
                       local structures. Large values sacrifice local details
                       for a broader global embedding.           [default: 15]
  --p-min-dist NUMBER  Controls the cluster size. Low values cause clumpier
    Range(0, None)     clusters. Higher values preserve a broad topological
                       structure. To get less overlapping data points the
                       default value is set to 0.4. For more details visit:
                       https://umap-learn.readthedocs.io/en/latest/parameters.
                       html                                     [default: 0.4]
Outputs:
  --o-umap ARTIFACT    The resulting UMAP matrix.
    PCoAResults                                                     [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.methods import umap

Docstring:

Uniform Manifold Approximation and Projection

Apply Uniform Manifold Approximation and Projection.

Parameters
----------
distance_matrix : DistanceMatrix
    The distance matrix on which UMAP should be computed.
number_of_dimensions : Int % Range(2, None), optional
    Dimensions to reduce the distance matrix to.
n_neighbors : Int % Range(1, None), optional
    Provide the balance between local and global structure. Low values
    prioritize the preservation of local structures. Large values sacrifice
    local details for a broader global embedding.
min_dist : Float % Range(0, None), optional
    Controls the cluster size. Low values cause clumpier clusters. Higher
    values preserve a broad topological structure. To get less overlapping
    data points the default value is set to 0.4. For more details visit:
    https://umap-learn.readthedocs.io/en/latest/parameters.html

Returns
-------
umap : PCoAResults
    The resulting UMAP matrix.