<?xml version="1.0" encoding="utf-8"?>
<tool id="anndata_ops" name="AnnData Operations" version="0.0.1+galaxy2" profile="@PROFILE@">
  <description>modifies metadata and flags genes</description>
  <macros>
    <import>scanpy_macros2.xml</import>
  </macros>
  <expand macro="requirements"/>
  <command detect_errors="exit_code"><![CDATA[
ln -s '${input_obj_file}' input.h5 &&
python $operations
]]></command>
  <configfiles>
    <configfile name="operations">
import scanpy as sc
import logging

adata = sc.read('input.h5')

gene_name = '${gene_symbols_field}'
qc_vars = list()

#for $i, $s in enumerate($modifications)
adata.obs['${s.to_obs}'] = adata.obs['${s.from_obs}']
#end for

gene_names = getattr(adata.var, gene_name)

#for $i, $flag in enumerate($gene_flags)
k_cat = gene_names.str.startswith('${flag.startswith}')
if k_cat.sum() > 0:
    adata.var['${flag.flag}'] = k_cat
    qc_vars.append('${flag.flag}')
else:
    logging.warning('No genes starting with {} found, skip calculating expression of {} genes'.format('${flag.startswith}', '${flag.flag}'))
#end for


if len(qc_vars) > 0:
    pct_top = [${top_genes}]
    sc.pp.calculate_qc_metrics(adata, qc_vars=qc_vars, percent_top=pct_top, inplace=True)

if 'n_genes' not in adata.obs.columns:
    sc.pp.filter_cells(adata, min_genes=0)
if 'n_counts' not in adata.obs.columns:
    sc.pp.filter_cells(adata, min_counts=0)
if 'n_cells' not in adata.var.columns:
    sc.pp.filter_genes(adata, min_cells=0)
if 'n_counts' not in adata.var.columns:
    sc.pp.filter_genes(adata, min_counts=0)

adata.write('output.h5', compression='gzip')
    </configfile>
</configfiles>

  <inputs>
    <param name="input_obj_file" argument="input-object-file" type="data" format="h5,h5ad" label="Input object in hdf5 AnnData format"/>
    <expand macro="output_object_params_no_loom"/>
    <repeat name="modifications" title="Change field names in AnnData observations" min="0">
      <param name="from_obs" type="text" label="Original name" help="Name in observations that you want to change">
        <sanitizer>
          <valid initial="string.printable"/>
        </sanitizer>
      </param>
      <param name="to_obs" type="text" label="New name" help="New name in observations that you want to change"/>
    </repeat>
    <param name="gene_symbols_field" value='index' type="text" label="Gene symbols field in AnnData" help="Field inside var.params where the gene symbols are, normally 'index' or 'gene_symbols'"/>
    <repeat name="gene_flags" title="Flag genes that start with these names">
      <param name="startswith" type="text" label="Starts with" help="Text that you expect the genes to be flagged to start with, such as 'MT-' for mito genes"/>
      <param name="flag" type="text" label="Var name" help="Name of the column in var.names where this boolean flag is stored, for example 'mito' for mitochondrial genes."/>
    </repeat>
    <param name="top_genes" label="Number of top genes" value='50' help="to calculate percentage of the flagged genes in that number of top genes. Used by sc.pp.calculate_qc_metrics (integer)." type="integer"/>
  </inputs>

  <outputs>
    <expand macro="output_data_obj_no_loom" description="metadata changes on"/>
  </outputs>

  <tests>
    <test>
      <param name="input_obj_file" value="find_cluster.h5"/>
      <param name="input_format" value="anndata"/>
      <param name="color_by" value="louvain"/>
      <output name="output_h5ad" file="output.h5" ftype="h5" compare="sim_size"/>
    </test>
  </tests>

  <help><![CDATA[
=============================
Operations on AnnData objects
=============================

Performs the following operations:

* Change observation fields, mostly for downstreaming processes convenience. Multiple fields can be changed as one.
* Flag genes that start with a certain text: useful for flagging mitochondrial, spikes or other groups of genes.
* For the flags created, calculates qc metrics (pct_<flag>_counts).
* Calculates `n_genes`, `n_counts` for cells and `n_cells`, `n_counts` for genes.
* For top <N> genes specified, calculate qc metrics (pct_counts_in_top_<N>_genes).

This functionality will probably be added in the future to a larger package.
]]></help>
  <expand macro="citations"/>
</tool>