ugbio-mrd 1.2.0__tar.gz

ugbio_mrd-1.2.0/PKG-INFO

@@ -0,0 +1,15 @@
+Metadata-Version: 2.1
+Name: ugbio_mrd
+Version: 1.2.0
+Summary: Ultima Genomics MRD utils
+Author-email: Itai Rusinek <itai.rusinek@ultimagen.com>, Gat Krieger <gat.krieger@ultimagen.com>, Avigail Moldovan <avigail.moldovan@ultimagen.com>
+License: Apache-2.0
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: ugbio_core
+Requires-Dist: ugbio_ppmseq
+Requires-Dist: ugbio_featuremap
+
+# ugbio_mrd
+
+This module includes MRD (Minimal Residual Disease) python scripts and utils for bioinformatics pipelines.

ugbio_mrd-1.2.0/README.mrd.md

@@ -0,0 +1,3 @@
+# ugbio_mrd
+
+This module includes MRD (Minimal Residual Disease) python scripts and utils for bioinformatics pipelines.

ugbio_mrd-1.2.0/pyproject.toml

@@ -0,0 +1,40 @@
+[project]
+name = "ugbio_mrd"
+version = "1.2.0"
+requires-python = ">=3.10"
+dependencies = [
+    "ugbio_core",
+    "ugbio_ppmseq",
+    "ugbio_featuremap",
+]
+description = "Ultima Genomics MRD utils"
+authors = [
+    { name = "Itai Rusinek", email = "itai.rusinek@ultimagen.com" },
+    { name = "Gat Krieger", email = "gat.krieger@ultimagen.com" },
+    { name = "Avigail Moldovan", email = "avigail.moldovan@ultimagen.com" },
+]
+readme = "README.mrd.md"
+
+[project.license]
+text = "Apache-2.0"
+
+[project.scripts]
+run_tests = "pytest:main"
+generate_synthetic_signatures = "ugbio_mrd.generate_synthetic_signatures:main"
+intersect_featuremap_with_signature = "ugbio_mrd.intersect_featuremap_with_signature:main"
+prepare_data_from_mrd_pipeline = "ugbio_mrd.prepare_data_from_mrd_pipeline:main"
+
+[tool.uv.sources.ugbio_core]
+workspace = true
+
+[tool.uv.sources.ugbio_ppmseq]
+workspace = true
+
+[tool.uv.sources.ugbio_featuremap]
+workspace = true
+
+[build-system]
+requires = [
+    "setuptools>=61.0",
+]
+build-backend = "setuptools.build_meta"

ugbio_mrd-1.2.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

ugbio_mrd-1.2.0/tests/test_intersect_featuremap_with_signature.py

@@ -0,0 +1,33 @@
+import filecmp
+from os.path import join as pjoin
+from pathlib import Path
+
+import pytest
+from ugbio_mrd.mrd_utils import intersect_featuremap_with_signature
+
+
+@pytest.fixture
+def resources_dir():
+    return Path(__file__).parent / "resources"
+
+
+def test_intersect_featuremap_with_signature(tmpdir, resources_dir):
+    signature = pjoin(
+        resources_dir,
+        "150382-BC04.filtered_signature.chr22_12693463.vcf.gz",
+    )
+    featuremap = pjoin(
+        resources_dir,
+        "featuremap_150419-BC04.sorted.chr22_12693463.vcf.gz",
+    )
+    expected_intersection = pjoin(
+        resources_dir,
+        "featuremap_150419-BC04.sorted.chr22_12693463.intersection.vcf.gz",
+    )
+    output_intersection = pjoin(tmpdir, "intersection.vcf.gz")
+    intersect_featuremap_with_signature(
+        featuremap_file=featuremap,
+        signature_file=signature,
+        output_intersection_file=output_intersection,
+    )
+    filecmp.cmp(output_intersection, expected_intersection)

ugbio_mrd-1.2.0/tests/test_mrd_utils.py

@@ -0,0 +1,171 @@
+import subprocess
+from os.path import join as pjoin
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+import pytest
+from pandas.testing import assert_frame_equal
+
+from ugbio_mrd.mrd_utils import (
+    generate_synthetic_signatures,
+    intersect_featuremap_with_signature,
+    read_intersection_dataframes,
+    read_signature,
+)
+intersection_file_basename = "MRD_test_subsample.MRD_test_subsample_annotated_AF_vcf_gz_mrd_quality_snvs.intersection"
+
+
+@pytest.fixture
+def resources_dir():
+    return Path(__file__).parent / "resources"
+
+
+
+
+
+def _assert_read_signature(signature, expected_signature, expected_columns=None, possibly_null_columns=None):
+    expected_columns = expected_columns or [
+        "ref",
+        "alt",
+        "id",
+        "qual",
+        "af",
+    ]
+    possibly_null_columns = possibly_null_columns or [
+        "id",
+        "qual",
+    ]
+    for c in expected_columns:
+        assert c in signature.columns
+        if c not in possibly_null_columns:
+            assert not signature[c].isnull().all()
+            assert (signature[c] == expected_signature[c]).all() or np.allclose(signature[c], expected_signature[c])
+
+
+def test_read_signature_ug_mutect(tmpdir, resources_dir):
+    signature = read_signature(pjoin(resources_dir, "mutect_mrd_signature_test.vcf.gz"), return_dataframes=True)
+    signature_no_sample_name = read_signature(
+        pjoin(resources_dir, "mutect_mrd_signature_test.no_sample_name.vcf.gz"),
+        return_dataframes=True,
+    )  # make sure we can read the dataframe even if the sample name could not be deduced from the header
+    expected_signature = pd.read_hdf(pjoin(resources_dir, "mutect_mrd_signature_test.expected_output.h5"))
+    _assert_read_signature(
+        signature,
+        expected_signature,
+        expected_columns=[
+            "ref",
+            "alt",
+            "id",
+            "qual",
+            "af",
+            "depth_tumor_sample",
+            "cycle_skip_status",
+            "gc_content",
+            "left_motif",
+            "right_motif",
+            "mutation_type",
+        ],
+    )
+    _assert_read_signature(
+        signature_no_sample_name,
+        expected_signature,
+        expected_columns=[
+            "ref",
+            "alt",
+            "id",
+            "qual",
+            "af",
+            "depth_tumor_sample",
+            "cycle_skip_status",
+            "gc_content",
+            "left_motif",
+            "right_motif",
+            "mutation_type",
+        ],
+        possibly_null_columns=["id", "qual", "depth_tumor_sample", "af"],
+    )
+
+
+def test_read_signature_ug_dv(tmpdir, resources_dir):
+    signature = read_signature(pjoin(resources_dir, "dv_mrd_signature_test.vcf.gz"), return_dataframes=True)
+    expected_signature = pd.read_hdf(pjoin(resources_dir, "dv_mrd_signature_test.expected_output.h5"))
+    _assert_read_signature(
+        signature,
+        expected_signature,
+        expected_columns=[
+            "ref",
+            "alt",
+            "id",
+            "qual",
+            "af",
+            "depth_tumor_sample",
+            "cycle_skip_status",
+            "gc_content",
+            "left_motif",
+            "right_motif",
+            "mutation_type",
+        ],
+    )
+
+
+def test_read_signature_external(resources_dir):
+    signature = read_signature(pjoin(resources_dir, "external_somatic_signature.vcf.gz"), return_dataframes=True)
+    expected_signature = pd.read_hdf(pjoin(resources_dir, "external_somatic_signature.expected_output.h5"))
+
+    _assert_read_signature(signature, expected_signature)
+
+
+def test_intersect_featuremap_with_signature(tmpdir, resources_dir):
+    signature_file = pjoin(resources_dir, "Pa_46.FreshFrozen.chr20.70039_70995.vcf.gz")
+    featuremap_file = pjoin(resources_dir, "Pa_46.bsDNA.chr20_sample.vcf.gz")
+    test_file = pjoin(resources_dir, "intersected_featuremap.vcf.gz")
+
+    output_intersection_file = pjoin(tmpdir, "intersected.vcf.gz")
+    intersect_featuremap_with_signature(
+        featuremap_file,
+        signature_file,
+        output_intersection_file=output_intersection_file,
+    )
+    cmd1 = f"bcftools view -H {output_intersection_file}"
+    cmd2 = f"bcftools view -H {test_file}"
+    assert subprocess.check_output(cmd1, shell=True) == subprocess.check_output(cmd2, shell=True)
+
+
+def test_read_intersection_dataframes(tmpdir, resources_dir):
+    parsed_intersection_dataframe = read_intersection_dataframes(
+        pjoin(resources_dir, f"{intersection_file_basename}.expected_output.parquet"),
+        return_dataframes=True,
+    )
+    parsed_intersection_dataframe_expected = pd.read_parquet(
+        pjoin(resources_dir, f"{intersection_file_basename}.parsed.expected_output.parquet")
+    )
+    parsed_intersection_dataframe2 = read_intersection_dataframes(
+        [pjoin(resources_dir, f"{intersection_file_basename}.expected_output.parquet")],
+        return_dataframes=True,
+    )
+    assert_frame_equal(
+        parsed_intersection_dataframe.reset_index(),
+        parsed_intersection_dataframe_expected,
+    )
+    assert_frame_equal(
+        parsed_intersection_dataframe2.reset_index(),
+        parsed_intersection_dataframe_expected,
+    )
+
+
+def test_generate_synthetic_signatures(tmpdir, resources_dir):
+    signature_file = pjoin(resources_dir, "mutect_mrd_signature_test.vcf.gz")
+    db_file = pjoin(
+        resources_dir,
+        "pancan_pcawg_2020.mutations_hg38_GNOMAD_dbsnp_beds.sorted.Annotated.HMER_LEN.edited.chr19.vcf.gz",
+    )
+    synthetic_signature_list = generate_synthetic_signatures(
+        signature_vcf=signature_file, db_vcf=db_file, n_synthetic_signatures=1, output_dir=tmpdir
+    )
+    signature = read_signature(synthetic_signature_list[0], return_dataframes=True)
+    expected_signature = read_signature(pjoin(resources_dir, "synthetic_signature_test.vcf.gz"), return_dataframes=True)
+    # test that motif distribution is the same (0th order)
+    assert (
+        signature.groupby(["ref", "alt"]).value_counts() == expected_signature.groupby(["ref", "alt"]).value_counts()
+    ).all()

ugbio_mrd-1.2.0/ugbio_mrd/generate_synthetic_signatures.py

@@ -0,0 +1,84 @@
+#!/env/python
+# Copyright 2022 Ultima Genomics Inc.
+#
+#    Licensed under the Apache License, Version 2.0 (the "License");
+#    you may not use this file except in compliance with the License.
+#    You may obtain a copy of the License at
+#
+#        http://www.apache.org/licenses/LICENSE-2.0
+#
+#    Unless required by applicable law or agreed to in writing, software
+#    distributed under the License is distributed on an "AS IS" BASIS,
+#    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#    See the License for the specific language governing permissions and
+#    limitations under the License.
+# DESCRIPTION
+#    Generates multiple synthetic signatures from a database, with the same
+#    trinucleotide substitution context as the input signature.
+# CHANGELOG in reverse chronological order
+import argparse
+import sys
+
+from ugbio_mrd.mrd_utils import generate_synthetic_signatures
+
+
+def __parse_args(argv: list[str]) -> argparse.Namespace:
+    parser = argparse.ArgumentParser(prog="generate_synthetic_signature_from_db", description=run.__doc__)
+    parser.add_argument(
+        "-s",
+        "--signature_vcf",
+        type=str,
+        required=True,
+        help="""Signature vcf file""",
+    )
+    parser.add_argument(
+        "-db",
+        "--db_vcf",
+        type=str,
+        required=True,
+        help="""Database vcf file (for example, PCAWG)""",
+    )
+    parser.add_argument(
+        "-n",
+        "--n_synthetic_signatures",
+        type=int,
+        required=True,
+        help="""Number of synthetic signatures to generate""",
+    )
+    parser.add_argument(
+        "-r",
+        "--ref_fasta",
+        type=str,
+        required=False,
+        help="reference fasta file, default None. Required if input vcf is not annotated with left and right motifs "
+        "X_LM and X_RM",
+    )
+    parser.add_argument(
+        "-o",
+        "--output_dir",
+        type=str,
+        default=None,
+        required=False,
+        help="""Output directory for synthetic signatures""",
+    )
+    return parser.parse_args(argv[1:])
+
+
+def run(argv):
+    """Generates multiple synthetic signatures from a database,
+    with the same trinucleotide substitution context as the input signature"""
+    args_in = __parse_args(argv)
+    generate_synthetic_signatures(
+        signature_vcf=args_in.signature_vcf,
+        db_vcf=args_in.db_vcf,
+        n_synthetic_signatures=args_in.n_synthetic_signatures,
+        output_dir=args_in.output_dir,
+        ref_fasta=args_in.ref_fasta,
+    )
+
+def main():
+    run(sys.argv)
+
+
+if __name__ == "__main__":
+    main()

ugbio_mrd-1.2.0/ugbio_mrd/intersect_featuremap_with_signature.py

@@ -0,0 +1,75 @@
+#!/env/python
+# Copyright 2022 Ultima Genomics Inc.
+#
+#    Licensed under the Apache License, Version 2.0 (the "License");
+#    you may not use this file except in compliance with the License.
+#    You may obtain a copy of the License at
+#
+#        http://www.apache.org/licenses/LICENSE-2.0
+#
+#    Unless required by applicable law or agreed to in writing, software
+#    distributed under the License is distributed on an "AS IS" BASIS,
+#    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#    See the License for the specific language governing permissions and
+#    limitations under the License.
+# DESCRIPTION
+#    Intersects featuremap vcf-like with pre-defined signature VCF-like
+# CHANGELOG in reverse chronological order
+import argparse
+import sys
+
+from ugbio_mrd.mrd_utils import intersect_featuremap_with_signature
+
+
+def __parse_args(argv: list[str]) -> argparse.Namespace:
+    parser = argparse.ArgumentParser(prog="intersect_with_signature", description=run.__doc__)
+    parser.add_argument(
+        "-f",
+        "--featuremap",
+        type=str,
+        required=True,
+        help="""Featuremap vcf file""",
+    )
+    parser.add_argument(
+        "-s",
+        "--signature",
+        type=str,
+        required=True,
+        help="""Signature vcf file""",
+    )
+    parser.add_argument(
+        "-o",
+        "--output",
+        type=str,
+        default=None,
+        required=False,
+        help="""Output intersection vcf file (lines from featuremap propagated)""",
+    )
+    parser.add_argument(
+        "-stype",
+        "--signature_type",
+        type=str,
+        default=None,
+        required=False,
+        help="""matched, control or db_control""",
+    )
+    return parser.parse_args(argv[1:])
+
+
+def run(argv):
+    """Intersect featuremap and signature vcf files on position and matching ref and alts"""
+    args_in = __parse_args(argv)
+    intersect_featuremap_with_signature(
+        featuremap_file=args_in.featuremap,
+        signature_file=args_in.signature,
+        output_intersection_file=args_in.output,
+        signature_type=args_in.signature_type,
+    )
+
+
+def main():
+    run(sys.argv)
+
+
+if __name__ == "__main__":
+    main()