Merge pull request #72 from sbslee/0.38.0-dev

0.38.0 dev

Author: sbslee

CHANGELOG.rst

@@ -1,6 +1,12 @@
 Changelog
 *********
 
+0.38.0 (2024-06-16)
+-------------------
+
+* Update :meth:`pyvcf.has_chr_prefix` method to ignore the HLA contigs for GRCh38.
+* :issue:`71`: Deprecate :meth:`common.plot_cytobands` method.
+
 0.37.0 (2023-09-09)
 -------------------
 

fuc/api/common.py

@@ -22,7 +22,6 @@
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
-from matplotlib.collections import BrokenBarHCollection
 import matplotlib.patches as mpatches
 import seaborn as sns
 
@@ -841,138 +840,6 @@ def extract_sequence(fasta, region):
         sequence = ''
     return sequence
 
-def plot_cytobands(cytoband, bed, ax=None, figsize=None):
-    """
-    Create chromosome ideograms along with BED data.
-
-    The method's source code is derived from a Python script (ideograms.py)
-    written by Ryan Dale. The original script can be found at:
-    https://gist.github.com/daler/c98fc410282d7570efc3#file-ideograms-py
-
-    Parameters
-    ----------
-    cytoband : str
-        Text file containing cytoband ideogram information.
-    bed : str
-        BED file to be displayed.
-    ax : matplotlib.axes.Axes, optional
-        Pre-existing axes for the plot. Otherwise, crete a new one.
-    figsize : tuple, optional
-        Width, height in inches. Format: (float, float).
-
-    Examples
-    --------
-
-    .. plot::
-        :context: close-figs
-
-        >>> import matplotlib.pyplot as plt
-        >>> from fuc import common
-        >>> common.load_dataset('cytoband')
-        >>> cytoband_file = '~/fuc-data/cytoband/cytoBandIdeo.txt.gz'
-        >>> bed_file = '~/fuc-data/cytoband/ucsc_genes.bed.gz'
-        >>> common.plot_cytobands(cytoband_file, bed_file, figsize=(10, 8))
-    """
-    def chromosome_collections(df, y_positions, height, **kwargs):
-        del_width = False
-        if 'width' not in df.columns:
-            del_width = True
-            df['width'] = df['end'] - df['start']
-        for chrom, group in df.groupby('chrom'):
-            yrange = (y_positions[chrom], height)
-            xranges = group[['start', 'width']].values
-            yield BrokenBarHCollection(
-                xranges, yrange, edgecolors=("black",), facecolors=group['colors'], **kwargs)
-        if del_width:
-            del df['width']
-
-    # Height of each ideogram
-    chrom_height = 1
-
-    # Spacing between consecutive ideograms
-    chrom_spacing = 1
-
-    # Height of the gene track. Should be smaller than `chrom_spacing` in order to
-    # fit correctly
-    gene_height = 0.4
-
-    # Padding between the top of a gene track and its corresponding ideogram
-    gene_padding = 0.1
-
-    # Decide which chromosomes to use
-    chromosome_list = [f'chr{i}' for i in list(range(1, 23)) + ['M', 'X', 'Y']]
-
-    # Keep track of the y positions for ideograms and genes for each chromosome,
-    # and the center of each ideogram (which is where we'll put the ytick labels)
-    ybase = 0
-    chrom_ybase = {}
-    gene_ybase = {}
-    chrom_centers = {}
-
-    # Iterate in reverse so that items in the beginning of `chromosome_list` will
-    # appear at the top of the plot
-    for chrom in chromosome_list[::-1]:
-        chrom_ybase[chrom] = ybase
-        chrom_centers[chrom] = ybase + chrom_height / 2.
-        gene_ybase[chrom] = ybase - gene_height - gene_padding
-        ybase += chrom_height + chrom_spacing
-
-    # Read in ideogram.txt, downloaded from UCSC Table Browser
-    ideo = pd.read_table(
-        cytoband,
-        names=['chrom', 'start', 'end', 'name', 'gieStain']
-    )
-
-    # Filter out chromosomes not in our list
-    ideo = ideo[ideo.chrom.apply(lambda x: x in chromosome_list)]
-
-    # Add a new column for width
-    ideo['width'] = ideo.end - ideo.start
-
-    # Colors for different chromosome stains
-    color_lookup = {
-        'gneg': (1., 1., 1.),
-        'gpos25': (.6, .6, .6),
-        'gpos50': (.4, .4, .4),
-        'gpos75': (.2, .2, .2),
-        'gpos100': (0., 0., 0.),
-        'acen': (.8, .4, .4),
-        'gvar': (.8, .8, .8),
-        'stalk': (.9, .9, .9),
-    }
-
-    # Add a new column for colors
-    ideo['colors'] = ideo['gieStain'].apply(lambda x: color_lookup[x])
-
-    # Same thing for genes
-    genes = pd.read_table(
-        bed,
-        names=['chrom', 'start', 'end', 'name'],
-        usecols=range(4))
-    genes = genes[genes.chrom.apply(lambda x: x in chromosome_list)]
-    genes['width'] = genes.end - genes.start
-    genes['colors'] = '#2243a8'
-
-    if ax is None:
-        fig, ax = plt.subplots(figsize=figsize)
-
-    # Now all we have to do is call our function for the ideogram data...
-    for collection in chromosome_collections(ideo, chrom_ybase, chrom_height):
-        ax.add_collection(collection)
-
-    # ...and the gene data
-    for collection in chromosome_collections(
-        genes, gene_ybase, gene_height, alpha=0.5, linewidths=0
-    ):
-        ax.add_collection(collection)
-
-    # Axes tweaking
-    ax.set_yticks([chrom_centers[i] for i in chromosome_list])
-    ax.set_yticklabels(chromosome_list)
-    ax.axis('tight')
-
-    return ax
-
 def convert_file2list(fn):
     """
     Convert a text file to a list of filenames.

fuc/api/pyvcf.py

@@ -763,6 +763,8 @@ def has_chr_prefix(file, size=1000):
     Return True if all of the sampled contigs from a VCF file have the
     (annoying) 'chr' string.
 
+    For GRCh38, the HLA contigs will be ignored.
+
     Parameters
     ----------
     file : str
@@ -779,6 +781,8 @@ def has_chr_prefix(file, size=1000):
     vcf = VariantFile(file)
     for record in vcf.fetch():
         n += 1
+        if record.chrom.startswith('HLA'):
+            continue
         if 'chr' not in record.chrom:
             return False
         if n > size:

fuc/version.py

@@ -1 +1 @@
-__version__ = '0.37.0'
+__version__ = '0.38.0'