Don't store full VCF header in vcz, only non-redundant meta information

bio2zarr/vcf.py

@@ -6,6 +6,7 @@
 import math
 import pathlib
 import pickle
+import re
 import shutil
 import sys
 import tempfile
@@ -973,8 +974,19 @@ def num_fields(self):
 
     @property
     def root_attrs(self):
+        meta_information_pattern = re.compile("##([^=]+)=(.*)")
+        vcf_meta_information = []
+        for line in self.vcf_header.split("\n"):
+            match = re.fullmatch(meta_information_pattern, line)
+            if match:
+                key = match.group(1)
+                if key in ("contig", "FILTER", "INFO", "FORMAT"):
+                    # these fields are stored in Zarr arrays
+                    continue
+                value = match.group(2)
+                vcf_meta_information.append((key, value))
         return {
-            "vcf_header": self.vcf_header,
+            "vcf_meta_information": vcf_meta_information,
         }
 
     def iter_id(self, start, stop):

bio2zarr/vcz.py

@@ -604,7 +604,7 @@ def init(
         root = zarr.open(store=self.path, mode="a", **zarr_utils.ZARR_FORMAT_KWARGS)
         root.attrs.update(
             {
-                "vcf_zarr_version": "0.2",
+                "vcf_zarr_version": "0.4",
                 "source": f"bio2zarr-{provenance.__version__}",
             }
         )

tests/test_vcf_examples.py

@@ -3,7 +3,6 @@
 import re
 from unittest import mock
 
-import cyvcf2
 import numpy as np
 import numpy.testing as nt
 import pytest
@@ -55,9 +54,22 @@ def test_filters(self, ds):
             ],
         )
 
-    def test_header(self, ds):
-        vcf = cyvcf2.VCF(self.data_path)
-        assert ds.attrs["vcf_header"] == vcf.raw_header
+    def test_vcf_meta_information(self, ds):
+        assert ds.attrs["vcf_meta_information"] == [
+            ["fileformat", "VCFv4.0"],
+            ["fileDate", "20090805"],
+            ["source", "myImputationProgramV3.1"],
+            ["reference", "1000GenomesPilot-NCBI36"],
+            ["phasing", "partial"],
+            ["ALT", '<ID=DEL:ME:ALU,Description="Deletion of ALU element">'],
+            ["ALT", '<ID=CNV,Description="Copy number variable region">'],
+            ["bcftools_viewVersion", "1.11+htslib-1.11-4"],
+            [
+                "bcftools_viewCommand",
+                "view -O b sample.vcf.gz; Date=Tue Feb 27 14:41:07 2024",
+            ],
+            ["bcftools_viewCommand", "view sample.bcf; Date=Wed Mar 27 11:42:16 2024"],
+        ]
 
     def test_source(self, ds):
         assert ds.attrs["source"] == f"bio2zarr-{provenance.__version__}"