Merge pull request #31 from biocore/improved_position_plots

Allow region slicing

ci/conda_requirements.txt

@@ -1,4 +1,3 @@
-tqdm
 scipy
 pyarrow<16.0.0
 polars-u64-idx>=1.21.0

micov/__init__.py

@@ -1,3 +1,11 @@
 """micov: microbiome coverage."""
+
 from . import _version
-__version__ = _version.get_versions()['version']
+
+__version__ = _version.get_versions()["version"]
+# note: currently for use with duckdb. we cannot easily enforce threads for polars
+# as a specific environment variable must be set prior to the first import. it's
+# doable but will need some engineeering to do it correctly.'And, polars does not
+# currently have a way to limit memory use.
+THREADS = 1
+MEMORY = 8  # gb

micov/_constants.py

@@ -1,48 +1,57 @@
 import polars as pl
 
-COLUMN_GENOME_ID = 'genome_id'
+COLUMN_GENOME_ID = "genome_id"
 COLUMN_GENOME_ID_DTYPE = str
-COLUMN_SAMPLE_ID = 'sample_id'
+COLUMN_SAMPLE_ID = "sample_id"
 COLUMN_SAMPLE_ID_DTYPE = str
-COLUMN_START = 'start'
+COLUMN_START = "start"
 COLUMN_START_DTYPE = pl.UInt32
-COLUMN_STOP = 'stop'
+COLUMN_STOP = "stop"
 COLUMN_STOP_DTYPE = pl.UInt32
-COLUMN_READ_ID = 'read'
+COLUMN_READ_ID = "read"
 COLUMN_READ_ID_DTYPE = str
-COLUMN_FLAG = 'flag'
+COLUMN_FLAG = "flag"
 COLUMN_FLAG_DTYPE = int
-COLUMN_CIGAR = 'cigar'
+COLUMN_CIGAR = "cigar"
 COLUMN_CIGAR_DTYPE = str
-COLUMN_LENGTH = 'length'
-COLUMN_LENGTH_DTYPE = int
-COLUMN_TAXONOMY = 'taxonomy'
+COLUMN_LENGTH = "length"
+COLUMN_LENGTH_DTYPE = pl.UInt32
+COLUMN_TAXONOMY = "taxonomy"
 COLUMN_TAXONOMY_DTYPE = str
-COLUMN_COVERED = 'covered'
+COLUMN_COVERED = "covered"
 COLUMN_COVERED_DTYPE = pl.UInt32
-COLUMN_PERCENT_COVERED = 'percent_covered'
+COLUMN_PERCENT_COVERED = "percent_covered"
 COLUMN_PERCENT_COVERED_DTYPE = float
 
 ### should really probably just use a dataclass, and type annotations?
 
+
 class _SCHEMA:
     def __init__(self):
         self.dtypes_dict = dict(self.dtypes_flat)
         self.columns = tuple([c for c, _ in self.dtypes_flat])
 
 
 class _BED_COV_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_START, COLUMN_START_DTYPE),
-                   (COLUMN_STOP, COLUMN_STOP_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_START, COLUMN_START_DTYPE),
+        (COLUMN_STOP, COLUMN_STOP_DTYPE),
+    ]
+
+
 BED_COV_SCHEMA = _BED_COV_SCHEMA()
 
 
 class _BED_COV_WITH_SAMPLEID_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_START, COLUMN_START_DTYPE),
-                   (COLUMN_STOP, COLUMN_STOP_DTYPE),
-                   (COLUMN_SAMPLE_ID, COLUMN_SAMPLE_ID_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_START, COLUMN_START_DTYPE),
+        (COLUMN_STOP, COLUMN_STOP_DTYPE),
+        (COLUMN_SAMPLE_ID, COLUMN_SAMPLE_ID_DTYPE),
+    ]
+
+
 BED_COV_SAMPLEID_SCHEMA = _BED_COV_WITH_SAMPLEID_SCHEMA()
 
 
@@ -52,48 +61,72 @@ class _SAM_SUBSET_SCHEMA(_SCHEMA):
     # for binary coverage, we don't care about the flag, but we're parsing it
     # now so we can care in the future.
     column_indices = [0, 1, 2, 3, 5]
-    dtypes_flat = [(COLUMN_READ_ID, COLUMN_READ_ID_DTYPE),
-                   (COLUMN_FLAG, COLUMN_FLAG_DTYPE),
-                   (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_START, COLUMN_START_DTYPE),
-                   (COLUMN_CIGAR, COLUMN_CIGAR_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_READ_ID, COLUMN_READ_ID_DTYPE),
+        (COLUMN_FLAG, COLUMN_FLAG_DTYPE),
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_START, COLUMN_START_DTYPE),
+        (COLUMN_CIGAR, COLUMN_CIGAR_DTYPE),
+    ]
+
+
 SAM_SUBSET_SCHEMA = _SAM_SUBSET_SCHEMA()
 
 
 class _SAM_SUBSET_SCHEMA_PARSED(_SCHEMA):
-    dtypes_flat = [(COLUMN_READ_ID, COLUMN_READ_ID_DTYPE),
-                   (COLUMN_FLAG, COLUMN_FLAG_DTYPE),
-                   (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_START, COLUMN_START_DTYPE),
-                   (COLUMN_CIGAR, COLUMN_CIGAR_DTYPE),
-                   (COLUMN_STOP, COLUMN_STOP_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_READ_ID, COLUMN_READ_ID_DTYPE),
+        (COLUMN_FLAG, COLUMN_FLAG_DTYPE),
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_START, COLUMN_START_DTYPE),
+        (COLUMN_CIGAR, COLUMN_CIGAR_DTYPE),
+        (COLUMN_STOP, COLUMN_STOP_DTYPE),
+    ]
+
+
 SAM_SUBSET_SCHEMA_PARSED = _SAM_SUBSET_SCHEMA_PARSED()
 
 
 class _GENOME_LENGTH_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE),
+    ]
+
+
 GENOME_LENGTH_SCHEMA = _GENOME_LENGTH_SCHEMA()
 
 
 class _GENOME_TAXONOMY_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_TAXONOMY, COLUMN_TAXONOMY_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_TAXONOMY, COLUMN_TAXONOMY_DTYPE),
+    ]
+
+
 GENOME_TAXNOMY_SCHEMA = _GENOME_TAXONOMY_SCHEMA()
 
 
 class _GENOME_COVERAGE_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_COVERED, COLUMN_COVERED_DTYPE),
-                   (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE),
-                   (COLUMN_PERCENT_COVERED, COLUMN_PERCENT_COVERED_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_COVERED, COLUMN_COVERED_DTYPE),
+        (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE),
+        (COLUMN_PERCENT_COVERED, COLUMN_PERCENT_COVERED_DTYPE),
+    ]
+
+
 GENOME_COVERAGE_SCHEMA = _GENOME_COVERAGE_SCHEMA()
 
 
 class _GENOME_COVERAGE_WITH_SAMPLEID_SCHEMA(_SCHEMA):
-    dtypes_flat = [(COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
-                   (COLUMN_COVERED, COLUMN_COVERED_DTYPE),
-                   (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE),
-                   (COLUMN_PERCENT_COVERED, COLUMN_PERCENT_COVERED_DTYPE),
-                   (COLUMN_SAMPLE_ID, COLUMN_SAMPLE_ID_DTYPE)]
+    dtypes_flat = [
+        (COLUMN_GENOME_ID, COLUMN_GENOME_ID_DTYPE),
+        (COLUMN_COVERED, COLUMN_COVERED_DTYPE),
+        (COLUMN_LENGTH, COLUMN_LENGTH_DTYPE),
+        (COLUMN_PERCENT_COVERED, COLUMN_PERCENT_COVERED_DTYPE),
+        (COLUMN_SAMPLE_ID, COLUMN_SAMPLE_ID_DTYPE),
+    ]
+
+
 GENOME_COVERAGE_WITH_SAMPLEID_SCHEMA = _GENOME_COVERAGE_WITH_SAMPLEID_SCHEMA()