<?xml version="1.0" encoding="utf-8"?>
<tool id="scanpy_run_umap" name="Scanpy RunUMAP" version="@TOOL_VERSION@+galaxy1">
  <description>visualise cell clusters using UMAP</description>
  <macros>
    <import>scanpy_macros2.xml</import>
  </macros>
  <expand macro="requirements"/>
  <command detect_errors="exit_code"><![CDATA[
ln -s '${input_obj_file}' input.h5 &&
PYTHONIOENCODING=utf-8 scanpy-run-umap
    --use-graph '${use_graph}'
    --key-added '${key_added}'
#if $embeddings
    --export-embedding embeddings.csv
#end if
#if $settings.default == "false"
    --n-components ${settings.n_components}
    --min-dist ${settings.min_dist}
    --spread ${settings.spread}
    --alpha ${settings.alpha}
    --gamma ${settings.gamma}
    --negative-sample-rate ${settings.negative_sample_rate}
    --random-state ${settings.random_seed}
    #if $settings.init_pos
        --init-pos '${settings.init_pos}'
    #end if
    #if $settings.maxiter
        --maxiter ${settings.maxiter}
    #end if
#end if
    @INPUT_OPTS@
    @OUTPUT_OPTS@

]]></command>

  <inputs>
    <expand macro="input_object_params"/>
    <expand macro="output_object_params"/>
    <param name="embeddings" type="boolean" checked="true" label="Output embeddings in csv format"/>
    <param name="use_graph" argument="--use-graph" value="neighbors" type="text"
           label="Name of the slot that holds the KNN graph"/>
    <param name="key_added" argument="--key-added" type="text" optional="true"
           label="Additional suffix to the name of the slot to save the embedding"/>

    <conditional name="settings">
      <param name="default" type="boolean" checked="true" label="Use programme defaults"/>
      <when value="true"/>
      <when value="false">
        <param name="n_components" argument="--n-components" type="integer" value="2" label="The number of dimensions of the embedding"/>
        <param name="min_dist" argument="--min-dist" type="float" value="0.5" label="The effective minimum distance between embedded points"/>
        <param name="spread" argument="--spread" type="float" value="1.0" label="The effective spread of embedded points"/>
        <param name="alpha" argument="--alpha" type="float" value="1.0" label="Initial learning rate"/>
        <param name="gamma" argument="--gamma" type="float" value="1.0" label="Weighting applied to negative samples"/>
        <param name="negative_sample_rate" argument="--negative-sample-rate" type="integer" value="5" label="The ratio of negative to positive edge in optimisation"/>
        <param name="init_pos" argument="--init-pos" type="text" label="Method to initialise embedding, any key for adata.obsm or choose from the preset methods">
          <option value="spectral" selected="true">spectral</option>
          <option value="paga">paga</option>
          <option value="random">random</option>
        </param>
        <param name="maxiter" argument="--maxiter" type="integer" optional="true" label="Number of iterations of optimisation"/>
        <param name="random_seed" argument="--random-state" type="integer" value="0" label="Seed for numpy random number generator"/>
      </when>
    </conditional>
  </inputs>

  <outputs>
    <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: UMAP object"/>
    <data name="output_embed" format="csv" from_work_dir="embeddings.csv" label="${tool.name} on ${on_string}: UMAP embeddings">
      <filter>embeddings</filter>
    </data>
  </outputs>

  <tests>
    <test>
      <param name="input_obj_file" value="find_cluster.h5"/>
      <param name="input_format" value="anndata"/>
      <param name="output_format" value="anndata"/>
      <param name="default" value="false"/>
      <param name="embeddings" value="true"/>
      <param name="random_seed" value="0"/>
      <output name="output_h5" file="run_umap.h5" ftype="h5" compare="sim_size"/>
      <output name="output_embed" file="run_umap.embeddings.csv" ftype="csv" compare="sim_size">
        <assert_contents>
          <has_n_columns n="2" sep=","/>
        </assert_contents>
      </output>
    </test>
  </tests>

  <help><![CDATA[
==========================================================
Embed the neighborhood graph using UMAP (`scanpy.tl.umap`)
==========================================================

For making UMAP plots, please use `Scanpy PlotEmbed` with the output of this tool and enter "umap" as the
name of the embedding to plot.

UMAP (Uniform Manifold Approximation and Projection) is a manifold learning
technique suitable for visualizing high-dimensional data. Besides tending to
be faster than tSNE, it optimizes the embedding such that it best reflects
the topology of the data, which we represent throughout Scanpy using a
neighborhood graph. tSNE, by contrast, optimizes the distribution of
nearest-neighbor distances in the embedding such that these best match the
distribution of distances in the high-dimensional space.  We use the
implementation of `umap-learn <https://github.com/lmcinnes/umap>`__
(McInnes et al, 2018). For a few comparisons of UMAP with tSNE, see this `preprint
<https://doi.org/10.1101/298430>`__.

It yields `X_umap`, UMAP coordinates of data.

@HELP@

@VERSION_HISTORY@
]]></help>
  <expand macro="citations"/>
</tool>