HTSeq 2.0.8__tar.gz → 2.0.9__tar.gz

htseq-2.0.9/HTSeq/_version.py

@@ -0,0 +1 @@
+__version__ = "2.0.9"

{htseq-2.0.8 → htseq-2.0.9}/HTSeq/features.py

@@ -239,8 +239,17 @@ def make_feature_dict(
     Args:
         feature_sequence (iterable of Feature): A sequence of features, e.g. as
             obtained from GFF_reader('myfile.gtf')
-        feature_type (string or None): If None, collect all features. If a
-            string, restrict to only one type of features, e.g. 'exon'.
+        feature_type (string, sequence of strings, or None): If None, collect
+            all features. If a string, restrict to only one type of features,
+            e.g. 'exon' (this is the most common situation). If a sequence of
+            strings, restrict to the types found in the sequence, e.g.
+            ['gene', 'pseudogene']. Using a feature of strings is an uncommon
+            need and can lead to a higher number of ambiguous alignments: only
+            use if you know what you are doing. Even then, beware that this
+            option is designed to work for feature types that are "peers" and
+            not obviously overlapping, such as genes and pseudogenes. If you
+            select nested features types (e.g. "gene" and "exon"), you are
+            likely to end up with meaningless numbers.
         feature_query (string or None): If None, all features of the selected
             types will be collected. If a string, it has to be in the format:
 
@@ -280,7 +289,7 @@ def make_feature_dict(
 
     features = {}
     for f in feature_sequence:
-        if feature_type in (None, f.type):
+        if any(ft in (None, f.type) for ft in feature_type):
             if f.type not in features:
                 features[f.type] = {}
             res_ftype = features[f.type]
@@ -322,8 +331,10 @@ def make_feature_genomicarrayofsets(
             attributes, separated by colons (:), will be used as an identifier.
             For instance, ['gene_id', 'exon_number'] uniquely identifies
             specific exons.
-        feature_type (string or None): If None, collect all features. If a
-            string, restrict to only one type of features, e.g. 'exon'.
+        feature_type (string, sequence of strings, or None): If None, collect
+            all features. If a string, restrict to only one type of features,
+            e.g. 'exon'. If a sequence of strings, restrict to the types found
+            in the sequence, e.g. 'gene' and 'pseudogene'
         feature_query (string or None): If None, all features of the selected
             types will be collected. If a string, it has to be in the format:
 
@@ -401,7 +412,7 @@ def make_feature_genomicarrayofsets(
     i = 0
     try:
         for f in feature_sequence:
-            if feature_type in (None, f.type):
+            if any(ft in (None, f.type) for ft in feature_type):
                 feature_id = get_id_attr(f, id_attribute)
 
                 if stranded and f.iv.strand == ".":

{htseq-2.0.8 → htseq-2.0.9}/HTSeq/scripts/count.py

@@ -176,6 +176,43 @@ def _check_samouts(sam_filenames, samout_format, samouts):
                         pass
 
 
+
+# Adapted from: https://github.com/python/cpython/issues/60603
+class OverwriteUniqueAppendAction(argparse.Action):
+    """Custom action to append unique values to a list, overwriting the default.
+
+    When using the `append` action, the default value is not removed
+    from the list. This problem is described in
+    https://github.com/python/cpython/issues/60603
+
+    This custom action aims to fix this problem by removing the default
+    value when the argument is specified for the first time.
+
+    Moreover, it only appends if the value is not already there, so the resulting
+    list has unique elements.
+    """
+
+    def __init__(self, option_strings, dest, nargs=None, **kwargs):
+        """Initialize the action."""
+        self.called_times = 0
+        self.default_value = kwargs.get("default")
+        super().__init__(option_strings, dest, **kwargs)
+
+    def __call__(self, parser, namespace, values, option_string=None):
+        """When the argument is specified on the commandline."""
+        current_values = getattr(namespace, self.dest)
+
+        if self.called_times == 0 and current_values == self.default_value:
+            current_values = []
+
+        # Only add if not already present (unique values)
+        if values not in current_values:
+            current_values.append(values)
+
+        setattr(namespace, self.dest, current_values)
+        self.called_times += 1
+
+
 def _parse_sanitize_cmdline_arguments():
     pa = argparse.ArgumentParser(
         add_help=False,
@@ -274,17 +311,23 @@ def _parse_sanitize_cmdline_arguments():
         "--type",
         type=str,
         dest="feature_type",
-        default="exon",
+        action=OverwriteUniqueAppendAction,
+        default=["exon"],
         help="Feature type (3rd column in GTF file) to be used, "
-        + "all features of other type are ignored (default, suitable for Ensembl "
-        + "GTF files: exon)",
+        + "all features of other type are ignored (default, suitable for"
+        + "Ensembl GTF files: exon). If you can call this option multiple times, "
+        + "features of all specified types will be included, e.g. to include "
+        + "both genes and pseudogenes you might use -t gene -t pseudogene. "
+        + "Calling this option multiple times is a rare need and might result "
+        + "in excessive numbers of ambiguous counts: only use if you know what "
+        + "you are doing.",
     )
     pa.add_argument(
         "-i",
         "--idattr",
         type=str,
         dest="idattr",
-        action="append",
+        action=OverwriteUniqueAppendAction,
         default=["gene_id"],
         help="GTF attribute to be used as feature ID (default, "
         + "suitable for Ensembl GTF files: gene_id). All feature of the "
@@ -299,7 +342,7 @@ def _parse_sanitize_cmdline_arguments():
     pa.add_argument(
         "--additional-attr",
         type=str,
-        action="append",
+        action='append',
         dest='additional_attributes',
         default=[],
         help="Additional feature attributes (default: none, "
@@ -354,7 +397,7 @@ def _parse_sanitize_cmdline_arguments():
         "--samout",
         type=str,
         dest="samouts",
-        action="append",
+        action='append',
         default=[],
         help="Write out all SAM alignment records into "
         + "SAM/BAM files (one per input file needed), annotating each line "
@@ -445,13 +488,6 @@ def _parse_sanitize_cmdline_arguments():
 
     args = pa.parse_args()
 
-    # Deal with custom id_attribute lists. This is never shorter than 1 because
-    # gene_id is the default. However, if the option was called at least once,
-    # that should _override_ the default, which means skipping the first
-    # element (i.e., gene_id).
-    if len(args.idattr) > 1:
-        del args.idattr[0]
-
     # Never use more CPUs than files
     args.nprocesses = min(args.nprocesses, len(args.samfilenames))
 

{htseq-2.0.8 → htseq-2.0.9}/HTSeq/scripts/count_with_barcodes.py

@@ -591,10 +591,13 @@ def main():
 
     pa.add_argument(
             "-t", "--type", type=str, dest="featuretype",
-            default="exon",
-            help="Feature type (3rd column in GTF file) to be used, " +
-            "all features of other type are ignored (default, suitable for Ensembl " +
-            "GTF files: exon)")
+            action="append", default=["exon"],
+            help="Feature type (3rd column in GTF file) to be used, all "
+            + "features of other type are ignored (default, suitable for"
+            + "Ensembl GTF files: exon). You can call this option multiple "
+            + "times. Features of all specified types will be included. "
+            + "E.g. to include both genes and pseudogenes you might use "
+            + "-t gene -t pseudogene")
 
     pa.add_argument(
             "-i", "--idattr", type=str, dest="idattr",

{htseq-2.0.8 → htseq-2.0.9}/HTSeq.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: HTSeq
-Version: 2.0.8
+Version: 2.0.9
 Summary: A framework to process and analyze data from high-throughput sequencing (HTS) assays
 Home-page: https://github.com/htseq
 Author: Simon Anders, Fabio Zanini

{htseq-2.0.8 → htseq-2.0.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: HTSeq
-Version: 2.0.8
+Version: 2.0.9
 Summary: A framework to process and analyze data from high-throughput sequencing (HTS) assays
 Home-page: https://github.com/htseq
 Author: Simon Anders, Fabio Zanini

htseq-2.0.9/VERSION

@@ -0,0 +1 @@
+2.0.9