comparison utils/gff_util.py @ 5:0145969324c4 draft

planemo upload for repository https://github.com/galaxyproject/tools-devteam/tree/master/tool_collections/gops/subtract commit cae3e05d02e60f595bb8b6d77a84f030e9bd1689

4:7a2a604ae9c8

"""
Provides utilities for working with GFF files.
"""

import copy
from bx.intervals.io import GenomicInterval, GenomicIntervalReader, MissingFieldError, NiceReaderWrapper
from bx.tabular.io import Header, Comment, ParseError
from utils.odict import odict


class GFFInterval( GenomicInterval ):
    """
    A GFF interval, including attributes. If file is strictly a GFF file,

        interval = GFFInterval( self, line.split( "\t" ), self.chrom_col, self.feature_col,
                                self.start_col, self.end_col, self.strand_col, self.score_col,
                                self.default_strand, fix_strand=self.fix_strand )
        return interval

    def next( self ):
        """ Returns next GFFFeature. """

        #
        # Helper function.
        #

        # intervals to read, this is where iterator dies.
        if not self.seed_interval:
            while not self.seed_interval:
                try:
                    self.seed_interval = GenomicIntervalReader.next( self )
                except ParseError, e:
                    handle_parse_error( e )
                # TODO: When no longer supporting python 2.4 use finally:
                #finally:
                raw_size += len( self.current_line )

        # If header or comment, clear seed interval and return it with its size.
        if isinstance( self.seed_interval, ( Header, Comment ) ):
            return_val = self.seed_interval

        feature_intervals.append( self.seed_interval )
        while True:
            try:
                interval = GenomicIntervalReader.next( self )
                raw_size += len( self.current_line )
            except StopIteration, e:
                # No more intervals to read, but last feature needs to be
                # returned.
                interval = None
                raw_size += len( self.current_line )
                break
            except ParseError, e:
                handle_parse_error( e )
                raw_size += len( self.current_line )
                continue
            # TODO: When no longer supporting python 2.4 use finally:
            #finally:
            #raw_size += len( self.current_line )

            # Ignore comments.
            if isinstance( interval, Comment ):
                continue

        # Convert to BED coords?
        if self.convert_to_bed_coord:
            convert_gff_coords_to_bed( feature )

        return feature


def convert_bed_coords_to_gff( interval ):
    """
    Converts an interval object's coordinates from BED format to GFF format.

    by transcript_id, chrom, and start position.
    """

    # -- Get function that generates line/feature key. --

    get_transcript_id = lambda fields: parse_gff_attributes( fields[8] )[ 'transcript_id' ]
    if strict:
        # Strict GTF parsing uses transcript_id only to group lines into feature.
        key_fn = get_transcript_id
    else:
        # Use lenient parsing where chromosome + transcript_id is the key. This allows
        # transcripts with same ID on different chromosomes; this occurs in some popular
        # datasources, such as RefGenes in UCSC.
        key_fn = lambda fields: fields[0] + '_' + get_transcript_id( fields )

    # Aggregate intervals by transcript_id and collect comments.
    feature_intervals = odict()
    comments = []
    for count, line in enumerate( iterator ):

5:0145969324c4

"""
Provides utilities for working with GFF files.
"""
import copy

from bx.intervals.io import GenomicInterval, GenomicIntervalReader, MissingFieldError, NiceReaderWrapper
from bx.tabular.io import Comment, Header, ParseError

from .odict import odict


class GFFInterval( GenomicInterval ):
    """
    A GFF interval, including attributes. If file is strictly a GFF file,

        interval = GFFInterval( self, line.split( "\t" ), self.chrom_col, self.feature_col,
                                self.start_col, self.end_col, self.strand_col, self.score_col,
                                self.default_strand, fix_strand=self.fix_strand )
        return interval

    def __next__( self ):
        """ Returns next GFFFeature. """

        #
        # Helper function.
        #

        # intervals to read, this is where iterator dies.
        if not self.seed_interval:
            while not self.seed_interval:
                try:
                    self.seed_interval = GenomicIntervalReader.next( self )
                except ParseError as e:
                    handle_parse_error( e )
                # TODO: When no longer supporting python 2.4 use finally:
                # finally:
                raw_size += len( self.current_line )

        # If header or comment, clear seed interval and return it with its size.
        if isinstance( self.seed_interval, ( Header, Comment ) ):
            return_val = self.seed_interval

        feature_intervals.append( self.seed_interval )
        while True:
            try:
                interval = GenomicIntervalReader.next( self )
                raw_size += len( self.current_line )
            except StopIteration as e:
                # No more intervals to read, but last feature needs to be
                # returned.
                interval = None
                raw_size += len( self.current_line )
                break
            except ParseError as e:
                handle_parse_error( e )
                raw_size += len( self.current_line )
                continue
            # TODO: When no longer supporting python 2.4 use finally:
            # finally:
            # raw_size += len( self.current_line )

            # Ignore comments.
            if isinstance( interval, Comment ):
                continue

        # Convert to BED coords?
        if self.convert_to_bed_coord:
            convert_gff_coords_to_bed( feature )

        return feature
    next = __next__  # This line should be removed once the bx-python port to Python3 is finished


def convert_bed_coords_to_gff( interval ):
    """
    Converts an interval object's coordinates from BED format to GFF format.

    by transcript_id, chrom, and start position.
    """

    # -- Get function that generates line/feature key. --

    def get_transcript_id(fields):
        return parse_gff_attributes( fields[8] )[ 'transcript_id' ]

    if strict:
        # Strict GTF parsing uses transcript_id only to group lines into feature.
        key_fn = get_transcript_id
    else:
        # Use lenient parsing where chromosome + transcript_id is the key. This allows
        # transcripts with same ID on different chromosomes; this occurs in some popular
        # datasources, such as RefGenes in UCSC.
        def key_fn(fields):
            return fields[0] + '_' + get_transcript_id( fields )

    # Aggregate intervals by transcript_id and collect comments.
    feature_intervals = odict()
    comments = []
    for count, line in enumerate( iterator ):