diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 75fef90b4531dae80fe7679230d11cf9744652ec..ea2f40c45968ed52baff0153146e607fa7781a05 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -3,7 +3,5 @@ build:
tags:
- docker
script:
- - pip install --upgrade Cython
- - pip install --upgrade numpy
- pip install --upgrade pip setuptools wheel
- pip install '.'
diff --git a/MANIFEST.in b/MANIFEST.in
new file mode 100644
index 0000000000000000000000000000000000000000..3b7a5d1f706eb94c1f4d36915c7514d2f47bf292
--- /dev/null
+++ b/MANIFEST.in
@@ -0,0 +1 @@
+include vtools/*.pyx
\ No newline at end of file
diff --git a/README.md b/README.md
index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..4a76129e544bd3c898c465a651d6e17a469d5471 100644
--- a/README.md
+++ b/README.md
@@ -0,0 +1,149 @@
+vtools
+======
+
+Little toolset operating over VCF files. Uses cyvcf2 and cython under
+the hood for speed.
+
+
+Tools
+-----
+
+### vtools-filter
+
+Filter VCF files based on a few criteria. Will output both a filtered VCF
+file, and a VCF file containing all the filtered-out variants.
+
+#### Filter criteria
+
+| name | meaning | optional |
+| ---- | ------- | -------- |
+| NON_CANONICAL | Non-canonical chromosome | Yes |
+| INDEX_UNCALLED | Index uncalled or homozygous reference | Yes |
+| TOO_HIGH_GONL_AF | Too high GonL allele frequency | Yes |
+| TOO_HIGH_GNOMAD_AF | Too high GnomAD allele frequency | Yes |
+| LOW_GQ | Too low GQ on index sample | Yes |
+| DELETED_ALLELE | The only ALT allele is a deleted allele | No |
+
+#### Configuration
+
+Configuration of filters goes by a little JSON file.
+See [here](cfg/example-filter.json) for an example.
+
+
+#### Usage
+
+```bash
+Usage: vtools-filter [OPTIONS]
+
+Options:
+ -i, --input PATH Path to input VCF file [required]
+ -o, --output PATH Path to output (filtered) VCF file
+ [required]
+ -t, --trash PATH Path to trash VCF file [required]
+ -p, --params-file PATH Path to filter params json [required]
+ --index-sample TEXT Name of index sample [required]
+ --immediate-return / --no-immediate-return
+ Immediately write filters to file upon
+ hitting one filter criterium. Default = True
+ --help Show this message and exit.
+
+```
+
+### vtools-stats
+
+Collects some general statistics about a VCF file, and writes a json to
+stdout.
+
+#### Usage
+
+```bash
+Usage: vtools-stats [OPTIONS]
+
+Options:
+ -i, --input FILE Input VCF file [required]
+ --help Show this message and exit.
+```
+
+### vtools-gcoverage
+
+Collect coverage metrics over a gVCF file for every exon or every transcript
+in a refFlat file. This assumes the input VCF file is at least similar to
+GATK's gVCF files. gVCF files are only expected to have one sample; if
+your input file contains multiple samples, we simply take the first only.
+
+Output is a simple TSV file with the following columns
+
+| column | meaning |
+| ------ | ------- |
+| exon | exon number |
+| gene | gene name / symbol / id |
+| mean_dp | mean DP value over the exon |
+| mean_gq | mean GQ value over the exon* |
+| median_dp | median DP value over the exon |
+| median_gq | median GQ value over the exon |
+| perc_at_least_{10, 20, 30, 50, 100}_dp | Percentage of exon with DP value over value |
+| perc_at_least_{10, 29, 30, 50, 90}_gq | Percentage of exon with GQ value over exon |
+| transcript | transcript name / symbol / id |
+
+*: mean GQ value is computed by first calculating the P-value of all GQ
+values, then calculating the mean over these P-values, and lastly
+converting this number back to a phred score.
+
+#### Usage
+
+```bash
+Usage: vtools-gcoverage [OPTIONS]
+
+Options:
+ -I, --input-gvcf PATH Path to input VCF file [required]
+ -R, --refflat-file PATH Path to refFlat file [required]
+ --per-exon / --per-transcript Collect metrics per exon or per transcript
+ --help Show this message and exit.
+```
+
+### vtools-evaluate
+
+Evaluate a VCF file to a baseline VCF file containing true positives.
+We only consider variants that are present in both VCF files. This makes
+it useful when the two VCF files have been produced by wildly different
+technologies. E.g, when comparing a WES VCF file vs a SNP array, this
+tool can be quite useful.
+
+Output is a simple JSON file listing counts of concordant and discordant
+alleles.
+
+Multisample VCF files are allowed; the samples to be evaluated have to be set
+through a CLI argument.
+
+
+#### Usage
+
+```bash
+Usage: vtools-evaluate [OPTIONS]
+
+Options:
+ -c, --call-vcf PATH Path to VCF with calls to be evaluated
+ [required]
+ -p, --positive-vcf PATH Path to VCF with known calls [required]
+ -cs, --call-samples TEXT Sample(s) in call-vcf to consider. May be
+ called multiple times [required]
+ -ps, --positive-samples TEXT Sample(s) in positive-vcf to consider. May be
+ called multiple times [required]
+ --help Show this message and exit.
+```
+
+## Installation
+
+* Python 3.6 at minimum
+* numpy and cython must be installed prior to installing vtools
+ * this will get fixed in the very near future
+
+After both requirements have been met, simply install vtools with
+
+```bash
+python setup.py install
+```
+
+## License
+
+MIT
diff --git a/cfg/example-filter.json b/cfg/example-filter.json
new file mode 100644
index 0000000000000000000000000000000000000000..0130356457567483466c33cc31fbc8593f807fea
--- /dev/null
+++ b/cfg/example-filter.json
@@ -0,0 +1,9 @@
+{
+ "canonical_chromosomes": true,
+ "gq_pass": 7,
+ "index_called": true,
+ "low_gnomad_af": 0.05,
+ "low_gonl_af": null,
+ "gnomad_vcf": "/path/to/vcf",
+ "gonl_vcf": null
+}
\ No newline at end of file
diff --git a/setup.py b/setup.py
index 7e18b07d2161d9d82b1ae727d79059b39f25efc8..4892ea274afbf40c28a2c01674b12164ec24ecca 100644
--- a/setup.py
+++ b/setup.py
@@ -7,20 +7,41 @@ setup.py
:license: MIT
"""
from os.path import abspath, dirname, join
+import sys
+import pkg_resources
+import subprocess
from setuptools import setup, find_packages
-try:
- from Cython.Build import cythonize
-except ImportError:
- raise NotImplementedError("Installing cython on the fly not yet supported")
+# Temporarily install dependencies required by setup.py before trying to
+# import them. From https://bitbucket.org/dholth/setup-requires
+sys.path[0:0] = ['setup-requires']
+pkg_resources.working_set.add_entry('setup-requires')
-try:
- import numpy as np
-except ImportError:
- raise NotImplementedError("Installing numpy on the fly not yet supported")
+def missing_requirements(specifiers):
+ for specifier in specifiers:
+ try:
+ pkg_resources.require(specifier)
+ except pkg_resources.DistributionNotFound:
+ yield specifier
+
+
+def install_requirements(specifiers):
+ to_install = list(specifiers)
+ if to_install:
+ cmd = [sys.executable, "-m", "pip", "install",
+ "-t", "setup-requires"] + to_install
+ subprocess.call(cmd)
+
+
+requires = ['cython', 'numpy']
+install_requirements(missing_requirements(requires))
+
+
+from Cython.Build import cythonize
+import numpy as np
readme_file = join(abspath(dirname(__file__)), "README.md")
with open(readme_file) as desc_handle:
@@ -32,13 +53,18 @@ for ext in cython_extensions:
ext.include_dirs.append(np.get_include())
setup(
- name="vtools",
- version="0.0.1",
+ name="v-tools",
+ version="1.0.0",
description="Various tools operating over VCF files",
+ long_description=long_desc,
author="Sander Bollen",
author_email="a.h.b.bollen@lumc.nl",
+ url="https://git.lumc.nl/klinische-genetica/capture-lumc/vtools",
license="MIT",
packages=find_packages(),
+ python_requires=">=3.6",
+ zip_safe=False,
+ include_package_data=True,
install_requires=[
"click",
"cyvcf2",
@@ -55,6 +81,10 @@ setup(
]
},
classifiers=[
+ "License :: OSI Approved :: MIT License",
+ "Programming Language :: Python :: 3 :: Only",
+ "Programming Language :: Python :: 3.6",
+ "Programming Language :: Python :: 3.7",
"Topic :: Scientific/Engineering :: Bio-Informatics"
],
ext_modules=cython_extensions
diff --git a/vtools/__init__.py b/vtools/__init__.py
index 3943b8bfe3980137859697c2aaecac96c527cf6d..fcd826dfa0dfee50e5362517fbb9f8d2a01e4ab5 100644
--- a/vtools/__init__.py
+++ b/vtools/__init__.py
@@ -7,4 +7,4 @@ vtools
:license: MIT
"""
-__version__ = '0.0.1'
+