Add samtools-depth-single process

[marcellevstek]

REID-2234

• Update CHANGELOG.rst for each commit separately:
  • Pay attention to write entries under the "Unreleased" section.
  • Mark all breaking changes as "BACKWARD INCOMPATIBLE:" and put them
    before non-breaking changes.
  • If a commit modifies a feature listed under "Unreleased" section,
    it might be sufficient to modify the existing CHANGELOG entry from previous
    commit(s).
• Bump the process version:
  • MAJOR version (first number): Backward incompatible changes (changes
    that break the api/interface). Examples: renaming the input/output, adding
    mandatory input, removing input/output...
  • MINOR version (middle number): add functionality or changes in a
    backwards-compatible manner. Examples: add output field, add non-mandatory
    input parameter, use a different tool that produces same results...
  • PATCH version (last number): changes/bug fixes that do not affect
    the api/interface. Examples: typo fix, change/add warning messages...
• All inputs are used in process.
• All output fields have a value assigned to them.

[romunov]

To avoid reading the entire file, you can limit the import to non-zero values, perhaps something along the lines of (untested):

import pandas as pd

chunkz = pd.read_csv(StringIO(stdout), sep="\t", names=["chrom", "pos", "depth"], chunksize=5000)
df = pd.concat((x.query("depth > 0") for x in chunkz))

This works, though:

from io import StringIO
import pandas as pd


file_name = "https://raw.githubusercontent.com/uiuc-cse/data-fa14/gh-pages/data/iris.csv"
df = pd.read_csv(file_name)

flow = StringIO()
xy = df.to_csv(flow)

chunkz = pd.read_csv(flow , chunksize=5)
pd.concat((x.query("species == 'setosa'") for x in chunkz))

Unnamed: 0	sepal_length	sepal_width	petal_length	petal_width	species
0	0	5.1	3.5	1.4	0.2	setosa
1	1	4.9	3.0	1.4	0.2	setosa
2	2	4.7	3.2	1.3	0.2	setosa
3	3	4.6	3.1	1.5	0.2	setosa
4	4	5.0	3.6	1.4	0.2	setosa
5	5	5.4	3.9	1.7	0.4	setosa
6	6	4.6	3.4	1.4	0.3	setosa
...

[jkokosar]

LGTM. In next iteration, consider adding additional output:

• HDF5 file format
• coverage data in BED file format + associated index for genome browsers (example, should be reimplemented for Python)