bio2zarr 0.1.4__tar.gz → 0.1.5__tar.gz

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/CHANGELOG.md

@@ -1,3 +1,9 @@
+# 0.1.5 2025-03-31
+
+- Add support for merging contig IDs across multiple VCFs (#335)
+
+- Add support for unindexed (and uncompressed) VCFs (#337)
+
 # 0.1.4 2025-03-10
 
 - Fix bug in handling all-missing genotypes (#328)

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: bio2zarr
-Version: 0.1.4
+Version: 0.1.5
 Summary: Convert bioinformatics data to Zarr
 Author-email: sgkit Developers <project@sgkit.dev>
 License:                                  Apache License
@@ -241,6 +241,7 @@ Requires-Dist: pytest-coverage; extra == "dev"
 Requires-Dist: pytest-xdist; extra == "dev"
 Requires-Dist: sgkit>=0.8.0; extra == "dev"
 Requires-Dist: tqdm; extra == "dev"
+Dynamic: license-file
 
 [![CI](https://github.com/sgkit-dev/bio2zarr/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/sgkit-dev/bio2zarr/actions/workflows/ci.yml)
 [![Coverage Status](https://coveralls.io/repos/github/sgkit-dev/bio2zarr/badge.svg)](https://coveralls.io/github/sgkit-dev/bio2zarr)

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/bio2zarr/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.1.4'
-__version_tuple__ = version_tuple = (0, 1, 4)
+__version__ = version = '0.1.5'
+__version_tuple__ = version_tuple = (0, 1, 5)

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/bio2zarr/cli.py

@@ -624,8 +624,8 @@ def vcfpartition(vcfs, verbose, num_partitions, partition_size):
         num_parts_per_path = max(1, num_partitions // len(vcfs))
 
     for vcf_path in vcfs:
-        indexed_vcf = vcf_utils.IndexedVcf(vcf_path)
-        regions = indexed_vcf.partition_into_regions(
+        vcf_file = vcf_utils.VcfFile(vcf_path)
+        regions = vcf_file.partition_into_regions(
             num_parts=num_parts_per_path, target_part_size=partition_size
         )
         for region in regions:

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/bio2zarr/vcf2zarr/icf.py

@@ -228,8 +228,8 @@ def fixed_vcf_field_definitions():
 
 
 def scan_vcf(path, target_num_partitions):
-    with vcf_utils.IndexedVcf(path) as indexed_vcf:
-        vcf = indexed_vcf.vcf
+    with vcf_utils.VcfFile(path) as vcf_file:
+        vcf = vcf_file.vcf
         filters = []
         pass_index = -1
         for h in vcf.header_iter():
@@ -270,10 +270,10 @@ def scan_vcf(path, target_num_partitions):
             filters=filters,
             fields=fields,
             partitions=[],
-            num_records=sum(indexed_vcf.contig_record_counts().values()),
+            num_records=sum(vcf_file.contig_record_counts().values()),
         )
 
-        regions = indexed_vcf.partition_into_regions(num_parts=target_num_partitions)
+        regions = vcf_file.partition_into_regions(num_parts=target_num_partitions)
         for region in regions:
             metadata.partitions.append(
                 VcfPartition(
@@ -324,14 +324,28 @@ def scan_vcfs(paths, show_progress, target_num_partitions, worker_processes=1):
     # are compatible.
     all_partitions = []
     total_records = 0
+    contigs = {}
     for metadata, _ in results:
         for partition in metadata.partitions:
             logger.debug(f"Scanned partition {partition}")
             all_partitions.append(partition)
+        for contig in metadata.contigs:
+            if contig.id in contigs:
+                if contig != contigs[contig.id]:
+                    raise ValueError(
+                        "Incompatible contig definitions: "
+                        f"{contig} != {contigs[contig.id]}"
+                    )
+            else:
+                contigs[contig.id] = contig
         total_records += metadata.num_records
         metadata.num_records = 0
         metadata.partitions = []
 
+    contig_union = list(contigs.values())
+    for metadata, _ in results:
+        metadata.contigs = contig_union
+
     icf_metadata, header = results[0]
     for metadata, _ in results[1:]:
         if metadata != icf_metadata:
@@ -1079,9 +1093,9 @@ class IntermediateColumnarFormatWriter:
             self.path,
             partition_index,
         ) as tcw:
-            with vcf_utils.IndexedVcf(partition.vcf_path) as ivcf:
+            with vcf_utils.VcfFile(partition.vcf_path) as vcf:
                 num_records = 0
-                for variant in ivcf.variants(partition.region):
+                for variant in vcf.variants(partition.region):
                     num_records += 1
                     last_position = variant.POS
                     tcw.append("CHROM", variant.CHROM)

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/bio2zarr/vcf_utils.py

@@ -89,7 +89,10 @@ class Region:
     end: Optional[int] = None
 
     def __post_init__(self):
-        if self.start is not None:
+        assert self.contig is not None
+        if self.start is None:
+            self.start = 1
+        else:
             self.start = int(self.start)
             assert self.start > 0
         if self.end is not None:
@@ -393,9 +396,12 @@ class VcfIndexType(Enum):
     TABIX = ".tbi"
 
 
-class IndexedVcf(contextlib.AbstractContextManager):
+class VcfFile(contextlib.AbstractContextManager):
     def __init__(self, vcf_path, index_path=None):
         self.vcf = None
+        self.file_type = None
+        self.index_type = None
+
         vcf_path = pathlib.Path(vcf_path)
         if not vcf_path.exists():
             raise FileNotFoundError(vcf_path)
@@ -408,30 +414,34 @@ class IndexedVcf(contextlib.AbstractContextManager):
                     vcf_path.suffix + VcfIndexType.CSI.value
                 )
                 if not index_path.exists():
-                    raise FileNotFoundError(
-                        f"Cannot find .tbi or .csi file for {vcf_path}"
-                    )
+                    # No supported index found
+                    index_path = None
         else:
             index_path = pathlib.Path(index_path)
+            if not index_path.exists():
+                raise FileNotFoundError(
+                    f"Specified index path {index_path} does not exist"
+                )
 
         self.vcf_path = vcf_path
         self.index_path = index_path
-        self.file_type = None
-        self.index_type = None
-
-        if index_path.suffix == VcfIndexType.CSI.value:
-            self.index_type = VcfIndexType.CSI
-        elif index_path.suffix == VcfIndexType.TABIX.value:
-            self.index_type = VcfIndexType.TABIX
-            self.file_type = VcfFileType.VCF
-        else:
-            raise ValueError("Only .tbi or .csi indexes are supported.")
+        if index_path is not None:
+            if index_path.suffix == VcfIndexType.CSI.value:
+                self.index_type = VcfIndexType.CSI
+            elif index_path.suffix == VcfIndexType.TABIX.value:
+                self.index_type = VcfIndexType.TABIX
+                self.file_type = VcfFileType.VCF
+            else:
+                raise ValueError("Only .tbi or .csi indexes are supported.")
 
         self.vcf = cyvcf2.VCF(vcf_path)
-        self.vcf.set_index(str(self.index_path))
+        if self.index_path is not None:
+            self.vcf.set_index(str(self.index_path))
+
         logger.debug(f"Loaded {vcf_path} with index {self.index_path}")
         self.sequence_names = None
 
+        self.index = None
         if self.index_type == VcfIndexType.CSI:
             # Determine the file-type based on the "aux" field.
             self.index = read_csi(self.index_path)
@@ -441,9 +451,17 @@ class IndexedVcf(contextlib.AbstractContextManager):
                 self.sequence_names = self.index.parse_vcf_aux()
             else:
                 self.sequence_names = self.vcf.seqnames
-        else:
+        elif self.index_type == VcfIndexType.TABIX:
             self.index = read_tabix(self.index_path)
+            self.file_type = VcfFileType.VCF
             self.sequence_names = self.index.sequence_names
+        else:
+            assert self.index is None
+            var = next(self.vcf)
+            self.sequence_names = [var.CHROM]
+            self.vcf.close()
+            # There doesn't seem to be a way to reset the iterator
+            self.vcf = cyvcf2.VCF(vcf_path)
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         if self.vcf is not None:
@@ -452,6 +470,8 @@ class IndexedVcf(contextlib.AbstractContextManager):
         return False
 
     def contig_record_counts(self):
+        if self.index is None:
+            return {self.sequence_names[0]: RECORD_COUNT_UNKNOWN}
         d = dict(zip(self.sequence_names, self.index.record_counts))
         if self.file_type == VcfFileType.BCF:
             d = {k: v for k, v in d.items() if v > 0}
@@ -460,12 +480,21 @@ class IndexedVcf(contextlib.AbstractContextManager):
     def count_variants(self, region):
         return sum(1 for _ in self.variants(region))
 
-    def variants(self, region):
-        start = 1 if region.start is None else region.start
-        for var in self.vcf(str(region)):
-            # Need to filter because of indels overlapping the region
-            if var.POS >= start:
+    def variants(self, region=None):
+        if self.index is None:
+            contig = self.sequence_names[0]
+            if region is not None:
+                assert region.contig == contig
+            for var in self.vcf:
+                if var.CHROM != contig:
+                    raise ValueError("Multi-contig VCFs must be indexed")
                 yield var
+        else:
+            start = 1 if region.start is None else region.start
+            for var in self.vcf(str(region)):
+                # Need to filter because of indels overlapping the region
+                if var.POS >= start:
+                    yield var
 
     def _filter_empty_and_refine(self, regions):
         """
@@ -505,6 +534,9 @@ class IndexedVcf(contextlib.AbstractContextManager):
             if target_part_size_bytes < 1:
                 raise ValueError("target_part_size must be positive")
 
+        if self.index is None:
+            return [Region(self.sequence_names[0])]
+
         # Calculate the desired part file boundaries
         file_length = os.stat(self.vcf_path).st_size
         if num_parts is not None:

{bio2zarr-0.1.4 → bio2zarr-0.1.5}/bio2zarr.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: bio2zarr
-Version: 0.1.4
+Version: 0.1.5
 Summary: Convert bioinformatics data to Zarr
 Author-email: sgkit Developers <project@sgkit.dev>
 License:                                  Apache License
@@ -241,6 +241,7 @@ Requires-Dist: pytest-coverage; extra == "dev"
 Requires-Dist: pytest-xdist; extra == "dev"
 Requires-Dist: sgkit>=0.8.0; extra == "dev"
 Requires-Dist: tqdm; extra == "dev"
+Dynamic: license-file
 
 [![CI](https://github.com/sgkit-dev/bio2zarr/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/sgkit-dev/bio2zarr/actions/workflows/ci.yml)
 [![Coverage Status](https://coveralls.io/repos/github/sgkit-dev/bio2zarr/badge.svg)](https://coveralls.io/github/sgkit-dev/bio2zarr)