Python HSPFragment.aln_annotation['similarity']方法代码示例

# 需要导入模块: from Bio.SearchIO._model import HSPFragment [as 别名]
# 或者: from Bio.SearchIO._model.HSPFragment import aln_annotation['similarity'] [as 别名]
    def _parse_hsp(self, root_hsp_frag_elem, query_id, hit_id):
        """Iterator that transforms Hit_hsps XML elements into HSP objects.

        :param root_hsp_frag_elem: the ``Hit_hsps`` tag
        :type root_hsp_frag_elem: XML element tag
        :param query_id: query ID
        :type query_id: string
        :param hit_id: hit ID
        :type hit_id: string

        """
        # Hit_hsps DTD:
        # <!ELEMENT Hsp (
        #        Hsp_num,
        #        Hsp_bit-score,
        #        Hsp_score,
        #        Hsp_evalue,
        #        Hsp_query-from,
        #        Hsp_query-to,
        #        Hsp_hit-from,
        #        Hsp_hit-to,
        #        Hsp_pattern-from?,
        #        Hsp_pattern-to?,
        #        Hsp_query-frame?,
        #        Hsp_hit-frame?,
        #        Hsp_identity?,
        #        Hsp_positive?,
        #        Hsp_gaps?,
        #        Hsp_align-len?,
        #        Hsp_density?,
        #        Hsp_qseq,
        #        Hsp_hseq,
        #        Hsp_midline?)>

        # if value is None, feed the loop below an empty list
        if root_hsp_frag_elem is None:
            root_hsp_frag_elem = []

        for hsp_frag_elem in root_hsp_frag_elem:
            coords = {}  # temporary container for coordinates
            frag = HSPFragment(hit_id, query_id)
            for key, val_info in _ELEM_FRAG.items():
                value = hsp_frag_elem.findtext(key)
                caster = val_info[1]

                # adjust 'from' and 'to' coordinates to 0-based ones
                if value is not None:
                    if key.endswith('-from') or key.endswith('-to'):
                        # store coordinates for further processing
                        coords[val_info[0]] = caster(value)
                        continue
                    # recast only if value is not intended to be str
                    elif caster is not str:
                        value = caster(value)
                    setattr(frag, val_info[0], value)

            # set the similarity characters into aln_annotation dict
            frag.aln_annotation['similarity'] = \
                    hsp_frag_elem.findtext('Hsp_midline')

            # process coordinates
            # since 'x-from' could be bigger than 'x-to', we need to figure
            # out which one is smaller/bigger since 'x_start' is always smaller
            # than 'x_end'
            for coord_type in ('query', 'hit', 'pattern'):
                start_type = coord_type + '_start'
                end_type = coord_type + '_end'
                try:
                    start = coords[start_type]
                    end = coords[end_type]
                except KeyError:
                    continue
                else:
                    # convert to python range and setattr
                    setattr(frag, start_type, min(start, end) - 1)
                    setattr(frag, end_type, max(start, end))

            # set alphabet, based on program
            prog = self._meta.get('program')
            if prog == 'blastn':
                frag.alphabet = generic_dna
            elif prog in ['blastp', 'blastx', 'tblastn', 'tblastx']:
                frag.alphabet = generic_protein

            hsp = HSP([frag])
            for key, val_info in _ELEM_HSP.items():
                value = hsp_frag_elem.findtext(key)
                caster = val_info[1]
                if value is not None:
                    if caster is not str:
                        value = caster(value)
                    setattr(hsp, val_info[0], value)
            # delete element after we finish parsing it
            hsp_frag_elem.clear()
            yield hsp

# 需要导入模块: from Bio.SearchIO._model import HSPFragment [as 别名]
# 或者: from Bio.SearchIO._model.HSPFragment import aln_annotation['similarity'] [as 别名]
    def __iter__(self):
        for rec in self.blast_iter:
            # set attributes to SearchIO's
            # get id and desc
            if rec.query.startswith('>'):
                rec.query = rec.query[1:]
            try:
                qid, qdesc = rec.query.split(' ', 1)
            except ValueError:
                qid, qdesc = rec.query, ''
            qdesc = qdesc.replace('\n', '').replace('\r', '')

            qresult = QueryResult(id=qid)
            qresult.program = rec.application.lower()
            qresult.target = rec.database
            qresult.seq_len = rec.query_letters
            qresult.version = rec.version

            # determine alphabet based on program
            if qresult.program == 'blastn':
                alphabet = generic_dna
            elif qresult.program in ['blastp', 'blastx', 'tblastn', 'tblastx']:
                alphabet = generic_protein

            # iterate over the 'alignments' (hits) and the hit table
            for idx, aln in enumerate(rec.alignments):
                # get id and desc
                if aln.title.startswith('> '):
                    aln.title = aln.title[2:]
                elif aln.title.startswith('>'):
                    aln.title = aln.title[1:]
                try:
                    hid, hdesc = aln.title.split(' ', 1)
                except ValueError:
                    hid, hdesc = aln.title, ''
                hdesc = hdesc.replace('\n', '').replace('\r', '')

                # iterate over the hsps and group them in a list
                hsp_list = []
                for bhsp in aln.hsps:
                    frag = HSPFragment(hid, qid)
                    frag.alphabet = alphabet
                    # set alignment length
                    frag.aln_span = bhsp.identities[1]
                    # set frames
                    try:
                        frag.query_frame = int(bhsp.frame[0])
                    except IndexError:
                        if qresult.program in ('blastp', 'tblastn'):
                            frag.query_frame = 0
                        else:
                            frag.query_frame = 1
                    try:
                        frag.hit_frame = int(bhsp.frame[1])
                    except IndexError:
                        if qresult.program in ('blastp', 'tblastn'):
                            frag.hit_frame = 0
                        else:
                            frag.hit_frame = 1
                    # set query coordinates
                    frag.query_start = min(bhsp.query_start,
                            bhsp.query_end) - 1
                    frag.query_end = max(bhsp.query_start, bhsp.query_end)
                    # set hit coordinates
                    frag.hit_start = min(bhsp.sbjct_start,
                            bhsp.sbjct_end) - 1
                    frag.hit_end = max(bhsp.sbjct_start, bhsp.sbjct_end)
                    # set query, hit sequences and its annotation
                    qseq = ''
                    hseq = ''
                    midline = ''
                    for seqtrio in zip(bhsp.query, bhsp.sbjct, bhsp.match):
                        qchar, hchar, mchar = seqtrio
                        if qchar == ' ' or hchar == ' ':
                            assert all(' ' == x for x in seqtrio)
                        else:
                            qseq += qchar
                            hseq += hchar
                            midline += mchar
                    frag.query, frag.hit = qseq, hseq
                    frag.aln_annotation['similarity'] = midline

                    # create HSP object with the fragment
                    hsp = HSP([frag])
                    hsp.evalue = bhsp.expect
                    hsp.bitscore = bhsp.bits
                    hsp.bitscore_raw = bhsp.score
                    # set gap
                    try:
                        hsp.gap_num = bhsp.gaps[0]
                    except IndexError:
                        hsp.gap_num = 0
                    # set identity
                    hsp.ident_num = bhsp.identities[0]
                    hsp.pos_num = bhsp.positives[0]
                    if hsp.pos_num is None:
                        hsp.pos_num = hsp[0].aln_span

                    hsp_list.append(hsp)

#.........这里部分代码省略.........