change bigwig track to use oxbow

Author: Nanguage

coolbox/core/coverage/highlights.py

@@ -16,7 +16,7 @@ class _Highlights(object):
     def fetch_data(self, gr: GenomeRange, **kwargs):
         gr = to_gr(gr)
         if gr.chrom not in list(self.interval_tree):
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
 
         return [
             (region.begin, region.end, region.data)

coolbox/core/coverage/vlines.py

@@ -10,7 +10,7 @@ def fetch_data(self, gr: GenomeRange):
         vlines_list = []
 
         if gr.chrom not in list(self.vlines_intval_tree):
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
 
         for region in sorted(self.vlines_intval_tree[gr.chrom][gr.start - 1:gr.end + 1]):
             vlines_list.append(region.begin)

coolbox/core/track/arcs/fetch.py

@@ -8,9 +8,9 @@ def fetch_intervals(self, bgz_file, gr: GenomeRange, gr2: GenomeRange = None) ->
         open_region = self.properties.get("open_region") == "yes"
         rows = list(pairix_query(bgz_file, gr, second=gr2, open_region=open_region, split=True))
         if not rows:
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
             if gr2:
-                gr2.change_chrom_names()
+                gr2 = gr2.change_chrom_names()
             rows = list(pairix_query(bgz_file, gr, second=gr2, open_region=open_region, split=True))
 
         return pd.DataFrame(rows)

coolbox/core/track/bed/fetch.py

@@ -20,7 +20,7 @@ def fetch_intervals(self, bgz_file, gr: GenomeRange) -> pd.DataFrame:
         """
         intervals, bed_type = self.load_range(bgz_file, gr)
         if len(intervals) == 0:
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
             intervals, bed_type = self.load_range(bgz_file, gr)
         if len(intervals) == 0:
             log.debug(f"No valid intervals were found in file {bgz_file} within range {gr}")

coolbox/core/track/gtf.py

@@ -98,7 +98,7 @@ def fetch_intervals(self, gr: GenomeRange):
         """
         rows = [row for row in tabix_query(self.bgz_file, gr.chrom, gr.start, gr.end)]
         if not rows:
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
             for row in tabix_query(self.bgz_file, gr.chrom, gr.start, gr.end):
                 rows.append(row)
 

coolbox/core/track/hist/bigwig.py

@@ -1,10 +1,10 @@
 import numpy as np
-import pandas as pd
 
 from coolbox.utilities import (
     split_genome_range, change_chrom_names,
     GenomeRange, get_logger, to_gr
 )
+import oxbow as ox
 from .base import HistBase
 
 log = get_logger(__name__)
@@ -39,13 +39,13 @@ def __init__(self, file, **kwargs):
             **kwargs
         })
         super().__init__(**properties)
-        import bbi
-        self.bw = bbi.open(self.properties['file'])
+        self.ds = ox.from_bigwig(self.properties['file'])
 
     def fetch_plot_data(self, gr: GenomeRange, **kwargs):
-        num_bins = self.get_num_bins()
         self.check_chrom_name(gr)
-        return self.fetch_scores(gr, num_bins)
+        intervals = self.fetch_data(gr)
+        values = intervals['value'].values
+        return values
 
     def fetch_data(self, gr: GenomeRange, **kwargs):
         """
@@ -59,47 +59,17 @@ def fetch_data(self, gr: GenomeRange, **kwargs):
             BigWig interval table.
         """
         chrom, start, end = split_genome_range(gr)
-        if chrom not in self.bw.chromsizes:
+        if chrom not in self.ds.chrom_names:
             chrom = change_chrom_names(chrom)
 
-        intervals = self.bw.fetch_intervals(chrom, start, end)
-        columns = list(intervals.columns)
-        if 'value' in columns:
-            columns[columns.index('value')] = 'score'
-        intervals.columns = columns
-
+        intervals = self.ds.regions(f"{chrom}:{start}-{end}").pd()
         return intervals
 
-    def get_num_bins(self, default_num=700):
-        num_bins = default_num
-        if 'number_of_bins' in self.properties:
-            try:
-                num_bins = int(self.properties['number_of_bins'])
-            except TypeError:
-                num_bins = default_num
-                log.warning("'number_of_bins' value: {} for bigwig file {} "
-                            "is not valid. Using default value (700)".format(self.properties['number_of_bins'],
-                                                                             self.properties['file']))
-        return num_bins
-
-    def fetch_scores(self, genome_range, num_bins, max_try_nums=5):
-        """Fetch bins scores within input chromosome range.
-        """
-        scores_per_bin = np.zeros(num_bins)
-        gr = to_gr(genome_range)
-        if gr.chrom not in self.bw.chromsizes:
-            gr.change_chrom_names()
-        try:
-            scores_per_bin = self.bw.fetch(gr.chrom, gr.start, gr.end, num_bins).astype(float)
-        except Exception as e:
-            log.warning(f"error found while reading bigwig scores: {e}")
-        return scores_per_bin
-
     def check_chrom_name(self, genome_range):
-        if genome_range.chrom not in self.bw.chromsizes:
-            genome_range.change_chrom_names()
+        if genome_range.chrom not in self.ds.chrom_names:
+            genome_range = genome_range.change_chrom_names()
 
-        if genome_range.chrom not in self.bw.chromsizes:
+        if genome_range.chrom not in self.ds.chrom_names:
             log.warning("Can not read region {} from bigwig file:\n\n"
                         "{}\n\nPlease check that the chromosome name is part of the bigwig file "
                         "and that the region is valid".format(str(genome_range), self.properties['file']))

coolbox/core/track/hist/snp.py

@@ -1,6 +1,3 @@
-import os.path as osp
-import subprocess as subp
-
 import numpy as np
 import pandas as pd
 
@@ -81,7 +78,7 @@ def fetch_data(self, gr: GenomeRange, **kwargs):
         ix_pval = self.properties['col_pval']
         rows = self.load_range(gr)
         if len(rows) == 0:
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
             rows = self.load_range(gr)
         df = pd.DataFrame(rows)
         if df.shape[0] > 0:

coolbox/core/track/ideogram.py

@@ -76,7 +76,7 @@ def lookup_band_color(self, band_type):
 
     def fetch_data(self, gr: GenomeRange, **kwargs):
         if gr.chrom not in self.interval_tree:
-            gr.change_chrom_names()
+            gr = gr.change_chrom_names()
         bands_in_region = sorted(self.interval_tree[gr.chrom][gr.start:gr.end])
         rows = []
         for itv in bands_in_region:

coolbox/utilities/genome.py

@@ -119,7 +119,7 @@ def parse_region_string(region_string):
             raise ValueError(f"Failure to parse region string {region_string}, please check that region format "
                              f"should be like \"chr:start-end\".")
 
-    def change_chrom_names(self):
+    def change_chrom_names(self) -> "GenomeRange":
         """
         >>> range1 = GenomeRange("chr1", 1000, 2000)
         >>> range1.chrom
@@ -131,7 +131,9 @@ def change_chrom_names(self):
         >>> range1.chrom
         'chr1'
         """
-        self.chrom = change_chrom_names(self.chrom)
+        new_chrom = change_chrom_names(self.chrom)
+        gr = GenomeRange(new_chrom, self.start, self.end)
+        return gr
 
     @property
     def length(self):

coolbox/utilities/hic/wrap.py

@@ -59,9 +59,9 @@ def fetch(self, gr1, gr2=None):
             gr2 = GenomeRange(gr2)
 
         if gr1.chrom.startswith("chr"):
-            gr1.change_chrom_names()
+            gr1 = gr1.change_chrom_names()
         if gr2.chrom.startswith("chr"):
-            gr2.change_chrom_names()
+            gr2 = gr2.change_chrom_names()
 
         binsize = self.infer_binsize(gr1)
         self.fetched_binsize = binsize  # expose fetched binsize
@@ -128,9 +128,9 @@ def fetch_pixels(self, genome_range1, genome_range2=None):
         if genome_range2 is None:
             genome_range2 = genome_range1
         if genome_range1.chrom.startswith("chr"):
-            genome_range1.change_chrom_names()
+            genome_range1 = genome_range1.change_chrom_names()
         if genome_range2.chrom.startswith("chr"):
-            genome_range2.change_chrom_names()
+            genome_range2 = genome_range2.change_chrom_names()
         binsize = self.infer_binsize(genome_range1)
         siter = self.__fetch_straw_iter(genome_range1, genome_range2, binsize)
         rows = [[i[0], i[1], i[2]] for i in siter]
@@ -305,9 +305,9 @@ def fetch(self, genome_range1, genome_range2=None):
         cool = self.get_cool(genome_range1)
 
         if genome_range1.chrom not in cool.chromnames:
-            genome_range1.change_chrom_names()
+            genome_range1 = genome_range1.change_chrom_names()
         if genome_range2.chrom not in cool.chromnames:
-            genome_range2.change_chrom_names()
+            genome_range2 = genome_range2.change_chrom_names()
 
         try:
             mat = cool.matrix(balance=self.balance).fetch(str(genome_range1), str(genome_range2))
@@ -325,9 +325,9 @@ def fetch_pixels(self, genome_range1, genome_range2=None, join=True):
             genome_range2 = genome_range1
 
         if genome_range1.chrom not in cool.chromnames:
-            genome_range1.change_chrom_names()
+            genome_range1 = genome_range1.change_chrom_names()
         if genome_range2.chrom not in cool.chromnames:
-            genome_range2.change_chrom_names()
+            genome_range2 = genome_range2.change_chrom_names()
 
         mat = cool.matrix(as_pixels=True, balance=self.balance, join=join)
         return mat.fetch(str(genome_range1), str(genome_range2))