Merge branch 'main' into bug_fix_#167

.github/workflows/ci.yml

@@ -3,8 +3,7 @@ name: CI
 on:
   push:
     branches: [ main ]
-    tags:
-      - "v*"  # Tag events matching v*, i.e. v1.0, v20.15.10
+
   pull_request:
     branches: [ main ]
 

.github/workflows/publish.yml

@@ -1,7 +1,8 @@
 name: Publish Python Package to PyPI
 
 on:
-  workflow_call:
+  release:
+    types: [published]
   workflow_dispatch:
 
 jobs:

CHANGES.md

@@ -1,8 +1,18 @@
 # Release notes
 
-## [Upcoming release](https://github.com/open2c/bioframe/compare/v0.6.2...HEAD)
+## [Upcoming release](https://github.com/open2c/bioframe/compare/v0.6.3...HEAD)
+
+## [v0.6.3](https://github.com/open2c/bioframe/compare/v0.6.2...v0.6.3)
+Date 2024-03-11
+
+Fixes:
+* Prevent dropout from `closest` in some cases of left intervals with no neighbors by @agalitsyna in https://github.com/open2c/bioframe/pull/185
+* Fix overlap returning float indexes causing failing tests (numpy v1.22.4, pandas v1.5.2)  by @agalitsyna in https://github.com/open2c/bioframe/pull/185
+
+**Full Changelog**: https://github.com/open2c/bioframe/compare/v0.6.2...v0.6.3
 
 ## [v0.6.2](https://github.com/open2c/bioframe/compare/v0.6.1...v0.6.2)
+Date 2024-02-08
 
 Changes:
 * cols and df_view_col passed to downstream functions by @smitkadvani in https://github.com/open2c/bioframe/pull/182

bioframe/__init__.py

@@ -1,6 +1,150 @@
-from ._version import __version__
-from .core import *
-from .io import *
-from .ops import *
-from .extras import *
-from .vis import *
+try:
+    from importlib.metadata import PackageNotFoundError, version
+except ImportError:
+    from importlib_metadata import PackageNotFoundError, version
+
+try:
+    __version__ = version("bioframe")
+except PackageNotFoundError:
+    __version__ = "unknown"
+
+__all__ = [
+    "arrops",
+    "from_any",
+    "from_dict",
+    "from_list",
+    "from_series",
+    "is_bedframe",
+    "is_cataloged",
+    "is_chrom_dtype",
+    "is_complete_ucsc_string",
+    "is_contained",
+    "is_covering",
+    "is_overlapping",
+    "is_sorted",
+    "is_tiling",
+    "is_viewframe",
+    "make_viewframe",
+    "parse_region",
+    "parse_region_string",
+    "sanitize_bedframe",
+    "to_ucsc_string",
+    "update_default_colnames",
+    "binnify",
+    "digest",
+    "frac_gc",
+    "frac_gene_coverage",
+    "frac_mapped",
+    "make_chromarms",
+    "pair_by_distance",
+    "seq_gc",
+    "SCHEMAS",
+    "UCSCClient",
+    "assemblies_available",
+    "assembly_info",
+    "fetch_centromeres",
+    "fetch_chromsizes",
+    "load_fasta",
+    "read_bam",
+    "read_bigbed",
+    "read_bigwig",
+    "read_chromsizes",
+    "read_pairix",
+    "read_tabix",
+    "read_table",
+    "to_bigbed",
+    "to_bigwig",
+    "assign_view",
+    "closest",
+    "cluster",
+    "complement",
+    "count_overlaps",
+    "coverage",
+    "expand",
+    "merge",
+    "overlap",
+    "select",
+    "select_indices",
+    "select_labels",
+    "select_mask",
+    "setdiff",
+    "sort_bedframe",
+    "subtract",
+    "trim",
+    "plot_intervals",
+    "to_ucsc_colorstring",
+]
+
+from .core import (
+    arrops,
+    from_any,
+    from_dict,
+    from_list,
+    from_series,
+    is_bedframe,
+    is_cataloged,
+    is_chrom_dtype,
+    is_complete_ucsc_string,
+    is_contained,
+    is_covering,
+    is_overlapping,
+    is_sorted,
+    is_tiling,
+    is_viewframe,
+    make_viewframe,
+    parse_region,
+    parse_region_string,
+    sanitize_bedframe,
+    to_ucsc_string,
+    update_default_colnames,
+)
+from .extras import (
+    binnify,
+    digest,
+    frac_gc,
+    frac_gene_coverage,
+    frac_mapped,
+    make_chromarms,
+    pair_by_distance,
+    seq_gc,
+)
+from .io import (
+    SCHEMAS,
+    UCSCClient,
+    assemblies_available,
+    assembly_info,
+    fetch_centromeres,
+    fetch_chromsizes,
+    load_fasta,
+    read_bam,
+    read_bigbed,
+    read_bigwig,
+    read_chromsizes,
+    read_pairix,
+    read_tabix,
+    read_table,
+    to_bigbed,
+    to_bigwig,
+)
+from .ops import (
+    assign_view,
+    closest,
+    cluster,
+    complement,
+    count_overlaps,
+    coverage,
+    expand,
+    merge,
+    overlap,
+    select,
+    select_indices,
+    select_labels,
+    select_mask,
+    setdiff,
+    sort_bedframe,
+    subtract,
+    trim,
+)
+from .vis import plot_intervals, to_ucsc_colorstring
+
+del version, PackageNotFoundError

bioframe/core/__init__.py

@@ -1,21 +1,50 @@
 from . import arrops
-
-from . import specs
-from .specs import *
-
-from . import stringops
-from .stringops import *
-
-from . import checks
-from .checks import *
-
-from . import construction
-from .construction import *
+from .checks import (
+    is_bedframe,
+    is_cataloged,
+    is_contained,
+    is_covering,
+    is_overlapping,
+    is_sorted,
+    is_tiling,
+    is_viewframe,
+)
+from .construction import (
+    from_any,
+    from_dict,
+    from_list,
+    from_series,
+    make_viewframe,
+    sanitize_bedframe,
+)
+from .specs import is_chrom_dtype, update_default_colnames
+from .stringops import (
+    is_complete_ucsc_string,
+    parse_region,
+    parse_region_string,
+    to_ucsc_string,
+)
 
 __all__ = [
     "arrops",
-    *specs.__all__,
-    *stringops.__all__,
-    *checks.__all__,
-    *construction.__all__,
+    "is_bedframe",
+    "is_cataloged",
+    "is_contained",
+    "is_covering",
+    "is_overlapping",
+    "is_sorted",
+    "is_tiling",
+    "is_viewframe",
+    "from_any",
+    "from_dict",
+    "from_list",
+    "from_series",
+    "make_viewframe",
+    "sanitize_bedframe",
+    "is_chrom_dtype",
+    "update_default_colnames",
+    "is_complete_ucsc_string",
+    "parse_region",
+    "parse_region_string",
+    "to_ucsc_string",
 ]

bioframe/core/arrops.py

@@ -264,6 +264,23 @@ def _overlap_intervals_legacy(starts1, ends1, starts2, ends2, closed=False, sort
 
     return overlap_ids
 
+def _convert_points_to_len1_segments(starts, ends):
+    """
+    Convert points to len1 segments for internal use in overlap().
+    This enables desired overlap behavior for points and preserves
+    behavior for semi-open intervals of len>=1.
+    Parameters
+    ----------
+    starts, ends : numpy.ndarray
+
+    Returns
+    -------
+    pseudo_ends : numpy.ndarray
+    An array of pseudo-ends for overlapping intervals.
+    """
+    pseudo_ends = ends.copy()
+    pseudo_ends[ends == starts] += 1
+    return [starts, pseudo_ends]
 
 def overlap_intervals(starts1, ends1, starts2, ends2, closed=False, sort=False):
     """
@@ -296,8 +313,11 @@ def overlap_intervals(starts1, ends1, starts2, ends2, closed=False, sort=False):
 
     starts1 = np.asarray(starts1)
     ends1 = np.asarray(ends1)
+    starts1, ends1 = _convert_points_to_len1_segments(starts1, ends1)
+
     starts2 = np.asarray(starts2)
     ends2 = np.asarray(ends2)
+    starts2, ends2 = _convert_points_to_len1_segments(starts2, ends2)
 
     # Concatenate intervals lists
     n1 = len(starts1)

bioframe/core/construction.py

@@ -156,8 +156,8 @@ def from_any(regions, fill_null=False, name_col="name", cols=None):
         raise ValueError(f"Unknown input format: {type(regions)}")
 
     if fill_null:
+        out_df[sk1] = pd.to_numeric(out_df[sk1]).fillna(0)
         try:
-            out_df[sk1].fillna(0, inplace=True)
             ends = []
             for i in range(len(out_df)):
                 if out_df[ek1].values[i] is None:

bioframe/extras.py

@@ -305,19 +305,17 @@ def _each(chrom_group):
         seq = str(seq[:])
         gc = []
         for _, bin in chrom_group.iterrows():
-            s = seq[bin.start : bin.end]
+            s = seq[bin["start"] : bin["end"]]
             gc.append(seq_gc(s, mapped_only=mapped_only))
         return gc
 
-    out = df.groupby("chrom", sort=False).apply(_each)
+    agg = df.groupby("chrom", sort=False)[["start", "end"]].apply(_each)
+    out_col = pd.Series(data=np.concatenate(agg.values), index=df.index).rename("GC")
 
     if return_input:
-        return pd.concat(
-            [df, pd.Series(data=np.concatenate(out), index=df.index).rename("GC")],
-            axis="columns",
-        )
+        return pd.concat([df, out_col], axis="columns")
     else:
-        return pd.Series(data=np.concatenate(out), index=df.index).rename("GC")
+        return out_col
 
 
 def seq_gc(seq, mapped_only=True):

bioframe/io/__init__.py

@@ -1,17 +1,34 @@
+from .assembly import assemblies_available, assembly_info
+from .fileops import (
+    load_fasta,
+    read_bam,
+    read_bigbed,
+    read_bigwig,
+    read_chromsizes,
+    read_pairix,
+    read_tabix,
+    read_table,
+    to_bigbed,
+    to_bigwig,
+)
+from .resources import UCSCClient, fetch_centromeres, fetch_chromsizes
 from .schemas import SCHEMAS
 
-from . import fileops
-from .fileops import *
-
-from . import resources
-from .resources import *
-
-from . import assembly
-from .assembly import *
-
 __all__ = [
+    "assemblies_available",
+    "assembly_info",
+    "read_table",
+    "read_chromsizes",
+    "read_tabix",
+    "read_pairix",
+    "read_bam",
+    "load_fasta",
+    "read_bigwig",
+    "to_bigwig",
+    "read_bigbed",
+    "to_bigbed",
+    "UCSCClient",
+    "fetch_centromeres",
+    "fetch_chromsizes",
     "SCHEMAS",
-    *fileops.__all__,
-    *resources.__all__,
-    *assembly.__all__,
 ]