biotite 0.41.1__cp310-cp310-macosx_10_16_arm64.whl

biotite/sequence/io/gff/__init__.py

@@ -0,0 +1,26 @@
+# This source code is part of the Biotite package and is distributed
+# under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
+# information.
+
+"""
+This subpackage is used for reading and writing sequence features in the
+*Generic Feature Format 3* (GFF3).
+
+It provides the :class:`GFFFile` class, a low-level line-based
+interface to this format, and high-level functions for extracting
+:class:`Annotation` objects.
+
+.. note: This package cannot create hierarchical data structures from
+   GFF 3 files. This means, that you cannot directly access the the
+   parent or child of a feature.
+   However, the ``Id`` and ``Name`` attributes are stored in the
+   qualifiers of the created :class:`Feature` objects. 
+   Hence, it is possible to implement such a data structure from this
+   information.
+"""
+
+__name__ = "biotite.sequence.io.gff"
+__author__ = "Patrick Kunzmann"
+
+from .file import *
+from .convert import *

biotite/sequence/io/gff/convert.py

@@ -0,0 +1,133 @@
+# This source code is part of the Biotite package and is distributed
+# under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
+# information.
+
+__name__ = "biotite.sequence.io.gff"
+__author__ = "Patrick Kunzmann"
+__all__ = ["get_annotation", "set_annotation"]
+
+from ...annotation import Location, Feature, Annotation
+
+
+def get_annotation(gff_file):
+    """
+    Parse a GFF3 file into an :class:`Annotation`.
+
+    The *type* column is used as the :attr:`Feature.key` attribute,
+    the locations (``loc``) are taken from the *start*, *end* and
+    *strand* columns and  the *attributes* column is parsed into the
+    :attr:`Feature.qual` attribute.
+    Multiple entries with the same ``ID`` attribute are interpreted
+    as the same feature.
+    Thus, for entries with the same ``ID``, the *type* and *attributes*
+    are only parsed once and the locations are aggregated from each
+    entry.
+    
+    Parameters
+    ----------
+    gff_file : GFFFile
+        The file tro extract the :class:`Annotation` object from.
+    
+    Returns
+    -------
+    annotation : Annotation
+        The extracted annotation.
+    """
+    annot = Annotation()
+    current_key = None
+    current_locs = None
+    current_qual = None
+    current_id = None
+    for _, _, type, start, end, _, strand, _, attrib in gff_file:
+        id = attrib.get("ID")
+        if id != current_id or id is None:
+            # current_key is None, when there is no previous feature
+            # (beginning of the file)
+            if current_key is not None:
+                # Beginning of new feature -> Save previous feature
+                annot.add_feature(
+                    Feature(current_key, current_locs, current_qual)
+                )
+            # Track new feature
+            current_key = type
+            current_locs = [Location(start, end, strand)]
+            current_qual = attrib
+        else:
+            current_locs.append(Location(start, end, strand))
+        current_id = id
+    # Save last feature
+    if current_key is not None:
+        annot.add_feature(Feature(current_key, current_locs, current_qual))
+    return annot
+
+
+def set_annotation(gff_file, annotation,
+                   seqid=None, source=None, is_stranded=True):
+    """
+    Write an :class:`Annotation` object into a GFF3 file.
+
+    Each feature will get one entry for each location it has.
+    :class:`Feature` objects with multiple locations require the ``ID``
+    qualifier in its :attr:`Feature.qual` attribute.
+    
+    Parameters
+    ----------
+    gff_file : GFFFile
+        The GFF3 file to write into.
+    annotation : Annotation
+        The annoation which is written to the GFF3 file.
+    seqid : str, optional
+        The content for the *seqid* column.
+    source : str, optional
+        The content for the *source* column.
+    is_stranded : bool, optional
+        If true, the strand of each feature is taken into account.
+        Otherwise the *strand* column is filled with '``.``'.
+    """
+    for feature in sorted(annotation):
+        if len(feature.locs) > 1 and "ID" not in feature.qual:
+            raise ValueError(
+                "The 'Id' qualifier is required "
+                "for features with multiple locations"
+            )
+        ## seqid ##
+        if seqid is not None and " " in seqid:
+            raise ValueError("The 'seqid' must not contain whitespaces")
+        ## source ##
+        #Nothing to be done
+        ## type ##
+        type = feature.key
+        ## strand ##
+        # Expect same strandedness for all locations
+        strand = list(feature.locs)[0].strand if is_stranded else None
+        ## score ##
+        score = None
+        ## attributes ##
+        attributes = feature.qual
+        # The previous properties are shared by all entries
+        # for this feature
+        # The following loop handles properties that change with each
+        # location
+        reverse_order = True if strand == Location.Strand.REVERSE else False
+        next_phase = 0
+        for loc in sorted(
+            feature.locs, key=lambda loc: loc.first, reverse=reverse_order
+        ):
+            ## start ##
+            start = loc.first
+            ## end ##
+            end = loc.last
+            ## strand ##
+            strand = loc.strand if is_stranded else None
+            ## phase ##
+            if type == "CDS":
+                phase = next_phase
+                # Subtract the length of the location
+                next_phase -= loc.last - loc.first + 1
+                next_phase %= 3
+            else:
+                phase = None
+            gff_file.append(
+                seqid, source, type, start, end,
+                score, strand, phase, attributes
+            )

biotite/sequence/io/gff/file.py

@@ -0,0 +1,434 @@
+# This source code is part of the Biotite package and is distributed
+# under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
+# information.
+
+__name__ = "biotite.sequence.io.gff"
+__author__ = "Patrick Kunzmann"
+__all__ = ["GFFFile"]
+
+import copy
+import string
+from urllib.parse import quote, unquote
+import warnings
+from ....file import TextFile, InvalidFileError
+from ...annotation import Location
+
+
+# All punctuation characters except
+# percent, semicolon, equals, ampersand, comma
+_NOT_QUOTED = "".join(
+    [char for char in string.punctuation if char not in "%;=&,"]
+) + " "
+
+
+class GFFFile(TextFile):
+    """
+    This class represents a file in *Generic Feature Format 3*
+    (`GFF3 <https://github.com/The-Sequence-Ontology/Specifications/blob/master/gff3.md>`_)
+    format.
+
+    Similar to GenBank files, GFF3 files contain information about
+    features of a reference sequence, but in a more concise and better
+    parsable way.
+    However, it does not provide additional meta information.
+
+    This class serves as low-level API for accessing GFF3 files.
+    It is used as a sequence of entries, where each entry is defined as
+    a non-comment and non-directive line.
+    Each entry consists of values corresponding to the 9 columns of
+    GFF3:
+
+    ==============  ===============================  ==========================================================
+    **seqid**       ``str``                          The ID of the reference sequence
+    **source**      ``str``                          Source of the data (e.g. ``Genbank``)
+    **type**        ``str``                          Type of the feature (e.g. ``CDS``)
+    **start**       ``int``                          Start coordinate of feature on the reference sequence
+    **end**         ``int``                          End coordinate of feature on the reference sequence
+    **score**       ``float`` or ``None``            Optional score (e.g. an E-value)
+    **strand**      ``Location.Strand`` or ``None``  Strand of the feature, ``None`` if feature is not stranded
+    **phase**       ``int`` or ``None``              Reading frame shift, ``None`` for non-CDS features
+    **attributes**  ``dict``                         Additional properties of the feature
+    ==============  ===============================  ==========================================================
+
+    Note that the entry index may not be equal to the line index,
+    because GFF3 files can contain comment and directive lines.
+
+    Notes
+    -----
+    Although the GFF3 specification allows mixing in reference sequence
+    data in FASTA format via the ``##FASTA`` directive, this class does
+    not support extracting the sequence information.
+    The content after the ``##FASTA`` directive is simply ignored.
+    Please provide the sequence via a separate file or read the FASTA
+    data directly via the :attr:`lines` attribute:
+    
+    >>> import os.path
+    >>> from io import StringIO
+    >>> gff_file = GFFFile.read(os.path.join(path_to_sequences, "indexing_test.gff3"))
+    >>> fasta_start_index = None
+    >>> for directive, line_index in gff_file.directives():
+    ...     if directive == "FASTA":
+    ...         fasta_start_index = line_index + 1
+    >>> fasta_data = StringIO("\\n".join(gff_file.lines[fasta_start_index:]))
+    >>> fasta_file = FastaFile.read(fasta_data)
+    >>> for seq_string in fasta_file.values():
+    ...     print(seq_string[:60] + "...")
+    TACGTAGCTAGCTGATCGATGTTGTGTGTATCGATCTAGCTAGCTAGCTGACTACACAAT...
+
+    Examples
+    --------
+    Reading and editing of an existing GFF3 file:
+
+    >>> import os.path
+    >>> gff_file = GFFFile.read(os.path.join(path_to_sequences, "gg_avidin.gff3"))
+    >>> # Get content of first entry
+    >>> seqid, source, type, start, end, score, strand, phase, attrib = gff_file[0]
+    >>> print(seqid)
+    AJ311647.1
+    >>> print(source)
+    EMBL
+    >>> print(type)
+    region
+    >>> print(start)
+    1
+    >>> print(end)
+    1224
+    >>> print(score)
+    None
+    >>> print(strand)
+    Strand.FORWARD
+    >>> print(phase)
+    None
+    >>> print(attrib)
+    {'ID': 'AJ311647.1:1..1224', 'Dbxref': 'taxon:9031', 'Name': 'Z', 'chromosome': 'Z', 'gbkey': 'Src', 'mol_type': 'genomic DNA'}
+    >>> # Edit the first entry: Simply add a score
+    >>> score = 1.0
+    >>> gff_file[0] = seqid, source, type, start, end, score, strand, phase, attrib
+    >>> # Delete first entry
+    >>> del gff_file[0]
+
+    Writing a new GFF3 file:
+
+    >>> gff_file = GFFFile()
+    >>> gff_file.append_directive("Example directive", "param1", "param2")
+    >>> gff_file.append(
+    ...     "SomeSeqID", "Biotite", "CDS", 1, 99,
+    ...     None, Location.Strand.FORWARD, 0,
+    ...     {"ID": "FeatureID", "product":"A protein"}
+    ... )
+    >>> print(gff_file)   #doctest: +NORMALIZE_WHITESPACE
+    ##gff-version 3
+    ##Example directive param1 param2
+    SomeSeqID   Biotite CDS     1       99      .       +       0       ID=FeatureID;product=A protein
+    """
+    
+    def __init__(self):
+        super().__init__()
+        # Maps entry indices to line indices
+        self._entries = None
+        # Stores the directives as (directive text, line index)-tuple
+        self._directives = None
+        # Stores whether the file has FASTA data
+        self._has_fasta = None
+        self._index_entries()
+        self.append_directive("gff-version", "3")
+    
+    @classmethod
+    def read(cls, file):
+        """
+        Read a GFF3 file.
+        
+        Parameters
+        ----------
+        file : file-like object or str
+            The file to be read.
+            Alternatively a file path can be supplied.
+        
+        Returns
+        -------
+        file_object : GFFFile
+            The parsed file.
+        """
+        file = super().read(file)
+        file._index_entries()
+        return file
+    
+    def insert(self, index, seqid, source, type, start, end,
+               score, strand, phase, attributes=None):
+        """
+        Insert an entry at the given index.
+        
+        Parameters
+        ----------
+        index : int
+            Index where the entry is inserted.
+            If the index is equal to the length of the file, the entry
+            is appended at the end of the file. 
+        seqid : str
+            The ID of the reference sequence.
+        source : str
+            Source of the data (e.g. ``Genbank``).
+        type : str
+            Type of the feature (e.g. ``CDS``).
+        start : int
+            Start coordinate of feature on the reference sequence.
+        end : int
+            End coordinate of feature on the reference sequence.
+        score : float or None
+            Optional score (e.g. an E-value).
+        strand : Location.Strand or None
+            Strand of the feature, ``None`` if feature is not stranded.
+        phase : int or None
+            Reading frame shift, ``None`` for non-CDS features.
+        attributes : dict, optional
+            Additional properties of the feature.
+        """
+        if index == len(self):
+            self.append(seqid, source, type, start, end,
+                        score, strand, phase, attributes)
+        else:
+            line_index = self._entries[index]
+            line = GFFFile._create_line(
+                seqid, source, type, start, end,
+                score, strand, phase, attributes
+            )
+            self.lines.insert(line_index, line)
+            self._index_entries()
+    
+    def append(self, seqid, source, type, start, end,
+               score, strand, phase, attributes=None):
+        """
+        Append an entry to the end of the file.
+        
+        Parameters
+        ----------
+        seqid : str
+            The ID of the reference sequence.
+        source : str
+            Source of the data (e.g. ``Genbank``).
+        type : str
+            Type of the feature (e.g. ``CDS``).
+        start : int
+            Start coordinate of feature on the reference sequence.
+        end : int
+            End coordinate of feature on the reference sequence.
+        score : float or None
+            Optional score (e.g. an E-value).
+        strand : Location.Strand or None
+            Strand of the feature, ``None`` if feature is not stranded.
+        phase : int or None
+            Reading frame shift, ``None`` for non-CDS features.
+        attributes : dict, optional
+            Additional properties of the feature.
+        """
+        if self._has_fasta:
+            raise NotImplementedError(
+                "Cannot append feature entries, "
+                "as this file contains additional FASTA data"
+            )
+        line = GFFFile._create_line(
+            seqid, source, type, start, end, score, strand, phase, attributes
+        )
+        self.lines.append(line)
+        # Fast update of entry index by adding last line
+        self._entries.append(len(self.lines) - 1)
+    
+    def append_directive(self, directive, *args):
+        """
+        Append a directive line to the end of the file.
+        
+        Parameters
+        ----------
+        directive : str
+            Name of the directive.
+        *args : str
+            Optional parameters for the directive.
+            Each argument is simply appended to the directive, separated
+            by a single space character.
+        
+        Raises
+        ------
+        NotImplementedError
+            If the ``##FASTA`` directive is used, which is not
+            supported.
+        
+        Examples
+        --------
+
+        >>> gff_file = GFFFile()
+        >>> gff_file.append_directive("Example directive", "param1", "param2")
+        >>> print(gff_file)
+        ##gff-version 3
+        ##Example directive param1 param2
+        """
+        if directive.startswith("FASTA"):
+            raise NotImplementedError(
+                "Adding FASTA information is not supported"
+            )
+        directive_line = "##" + directive + " " + " ".join(args)
+        self._directives.append((directive_line[2:], len(self.lines)))
+        self.lines.append(directive_line)
+    
+    def directives(self):
+        """
+        Get the directives in the file.
+        
+        Returns
+        -------
+        directives : list of tuple(str, int)
+            A list of directives, sorted by their line order.
+            The first element of each tuple is the name of the
+            directive (without ``##``), the second element is the index
+            of the corresponding line.
+        """
+        # Sort in line order
+        return sorted(self._directives, key=lambda directive: directive[1])
+        
+    def __setitem__(self, index, item):
+        seqid, source, type, start, end, score, strand, phase, attrib = item
+        line = GFFFile._create_line(
+            seqid, source, type, start, end, score, strand, phase, attrib
+        )
+        line_index = self._entries[index]
+        self.lines[line_index] = line
+
+    
+    def __getitem__(self, index):
+        if (index >= 0 and  index >= len(self)) or \
+           (index <  0 and -index >  len(self)):
+                raise IndexError(
+                    f"Index {index} is out of range for GFFFile with "
+                    f"{len(self)} entries"
+                )
+        
+        line_index = self._entries[index]
+        # Columns are tab separated
+        s = self.lines[line_index].strip().split("\t")
+        if len(s) != 9:
+            raise InvalidFileError(f"Expected 9 columns, but got {len(s)}")
+        seqid, source, type, start, end, score, strand, phase, attrib = s
+
+        seqid = unquote(seqid)
+        source = unquote(source)
+        type = unquote(type)
+        start = int(start)
+        end = int(end)
+        score = None if score == "." else float(score)
+        if strand == "+":
+            strand = Location.Strand.FORWARD
+        elif strand == "-":
+            strand = Location.Strand.REVERSE
+        else:
+            strand = None
+        phase = None if phase == "." else int(phase)
+        attrib = GFFFile._parse_attributes(attrib)
+
+        return seqid, source, type, start, end, score, strand, phase, attrib
+    
+    def __delitem__(self, index):
+        line_index = self._entries[index]
+        del self.lines[line_index]
+        self._index_entries()
+    
+    def __len__(self):
+        return len(self._entries)
+    
+    def _index_entries(self):
+        """
+        Parse the file for comment and directive lines.
+        Count these lines cumulatively, so that entry indices can be
+        mapped onto line indices.
+        Additionally track the line index of directive lines.
+        """
+        self._directives = []
+        # Worst case allocation -> all lines contain actual entries
+        self._entries = [None] * len(self.lines)
+        self._has_fasta = False
+        entry_counter = 0
+        for line_i, line in enumerate(self.lines):
+            if len(line) == 0 or line[0] == " ":
+                # Empty line -> do nothing
+                pass
+            elif line.startswith("#"):
+                # Comment or directive
+                if line.startswith("##"):
+                    # Directive
+                    # Omit the leading '##'
+                    self._directives.append((line[2:], line_i))
+                    if line[2:] == "FASTA":
+                        self._has_fasta = True
+                        # This parser does not support bundled FASTA
+                        # data
+                        warnings.warn(
+                            "Biotite does not support FASTA data mixed into "
+                            "GFF files, the FASTA data will be ignored"
+                        )
+                        # To ignore the following FASTA data, stop
+                        # parsing at this point
+                        break
+            else:
+                # Actual entry
+                self._entries[entry_counter] = line_i
+                entry_counter += 1
+        # Trim to correct size
+        self._entries = self._entries[:entry_counter]
+
+    @staticmethod
+    def _create_line(seqid, source, type, start, end,
+                     score, strand, phase, attributes):
+        """
+        Create a line for a newly created entry.
+        """
+        seqid = quote(seqid.strip(), safe=_NOT_QUOTED) \
+                if seqid is not None else "."
+        source = quote(source.strip(), safe=_NOT_QUOTED) \
+                 if source is not None else "."
+        type = type.strip()
+
+        # Perform checks
+        if len(seqid) == 0:
+            raise ValueError("'seqid' must not be empty")
+        if len(source) == 0:
+            raise ValueError("'source' must not be empty")
+        if len(type) == 0:
+            raise ValueError("'type' must not be empty")
+        if seqid[0] == ">":
+            raise ValueError("'seqid' must not start with '>'")
+        
+        score = str(score) if score is not None else "."
+        if strand == Location.Strand.FORWARD:
+            strand = "+"
+        elif strand == Location.Strand.REVERSE:
+            strand = "-"
+        else:
+            strand = "."
+        phase = str(phase) if phase is not None else "."
+        attributes = ";".join(
+            [quote(key, safe=_NOT_QUOTED) + "=" + quote(val, safe=_NOT_QUOTED)
+             for key, val in attributes.items()]
+        ) if attributes is not None and len(attributes) > 0 else "."
+
+        return "\t".join(
+            [seqid, source, type, str(start), str(end),
+             str(score), strand, phase, attributes]
+        )
+    
+    @staticmethod
+    def _parse_attributes(attributes):
+        """
+        Parse the *attributes* string into a dictionary.
+        """
+        if attributes == ".":
+            return {}
+
+        attrib_dict = {}
+        attrib_entries = attributes.split(";")
+        for entry in attrib_entries:
+            compounds = entry.split("=")
+            if len(compounds) != 2:
+                raise InvalidFileError(
+                    f"Attribute entry '{entry}' is invalid"
+                )
+            key, val = compounds
+            attrib_dict[unquote(key)] = unquote(val)
+        return attrib_dict

biotite/sequence/phylo/__init__.py

@@ -0,0 +1,36 @@
+# This source code is part of the Biotite package and is distributed
+# under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
+# information.
+
+"""
+This subpackage provides functions and data structures for creating
+(phylogenetic) trees.
+
+The :class:`Tree` is the central class in this subpackage.
+It wraps a *root* :class:`TreeNode` object.
+A :class:`TreeNode` is either an intermediate node, if it has child
+:class:`TreeNode` objects, or otherwise a leaf node.
+
+A :class:`Tree` is not a container itself:
+Objects, e.g species names or sequences, that are represented by the
+nodes, cannot be stored directly in a :class:`Tree` or
+:class:`TreeNode`.
+Instead, each leaf node has a reference index:
+These indices refer to a separate list or array, containing the actual
+reference objects.
+
+A :class:`Tree` can be created from or exported to a *Newick* notation,
+usingthe :func:`Tree.from_newick()` or :func:`Tree.to_newick()` method,
+respectively.
+
+A :class:`Tree` can be build from a pairwise distance matrix using the
+popular *UPGMA* (:func:`upgma()`) and *Neighbor-Joining*
+(:func:`neighbor_joining()`) algorithms.
+"""
+
+__name__ = "biotite.sequence.phylo"
+__author__ = "Patrick Kunzmann"
+
+from .tree import *
+from .upgma import *
+from .nj import *