Beispiel #1
0
    def _create_qresult(self, hit_blocks):
        """Create the Biopython data structures from the parsed data (PRIVATE)."""
        query_id = self.query_id
        hit_dict = OrderedDict()

        for output_index, block in enumerate(hit_blocks):
            hit_id = block['hit_id']

            frag = HSPFragment(hit_id, query_id)
            # frag.alphabet = generic_protein
            if block['query_start']:
                frag.query_start = block['query_start'] - 1
            else:
                frag.query_start = block['query_start']
            frag.query_end = block['query_end']
            if block['hit_start']:
                frag.hit_start = block['hit_start'] - 1
            else:
                frag.hit_start = block['hit_start']
            frag.hit_end = block['hit_end']
            frag.hit = block['hit_seq']
            frag.query = block['query_seq']

            hsp = HSP([frag])
            hsp.hit_id = hit_id
            hsp.output_index = output_index
            hsp.query_id = query_id
            hsp.hit_description = block['description']
            is_included = True  # Should everything should be included?
            hsp.is_included = is_included
            hsp.evalue = block['evalue']
            hsp.score = block['score']
            hsp.prob = block['prob']
            hsp.hit_seq_len = block['hit_seq_len']
            hsp.text = block['text']

            if hit_id not in hit_dict:
                hit = Hit([hsp], hit_id)
                hit.description = block['description']
                hit.is_included = is_included
                hit.evalue = block['evalue']
                hit.score = block['score']
                hit_dict[hit_id] = hit
            else:
                hit_dict[hit_id].append(hsp)

        qresult = QueryResult(hit_dict.values(), query_id)
        qresult.program = _PROGRAM
        qresult.seq_len = self.seq_len
        return [qresult]
Beispiel #2
0
    def _create_qresult(self, hit_blocks):
        """Create the Biopython data structures from the parsed data (PRIVATE)."""
        query_id = self.query_id
        hit_dict = OrderedDict()

        for output_index, block in enumerate(hit_blocks):
            hit_id = block["hit_id"]

            frag = HSPFragment(hit_id, query_id)
            frag.molecule_type = "protein"
            frag.query_start = block["query_start"] - 1
            frag.query_end = block["query_end"]
            frag.hit_start = block["hit_start"] - 1
            frag.hit_end = block["hit_end"]
            frag.hit = block["hit_seq"]
            frag.query = block["query_seq"]

            hsp = HSP([frag])
            hsp.hit_id = hit_id
            hsp.output_index = output_index
            hsp.query_id = query_id
            hsp.hit_description = block["description"]
            is_included = True  # Should everything should be included?
            hsp.is_included = is_included
            hsp.evalue = block["evalue"]
            hsp.score = block["score"]
            hsp.prob = block["prob"]

            if hit_id not in hit_dict:
                hit = Hit([hsp], hit_id)
                hit.description = block["description"]
                hit.is_included = is_included
                hit.evalue = block["evalue"]
                hit.score = block["score"]
                hit_dict[hit_id] = hit
            else:
                hit_dict[hit_id].append(hsp)

        qresult = QueryResult(hit_dict.values(), query_id)
        qresult.program = _PROGRAM
        qresult.seq_len = self.seq_len
        return [qresult]
Beispiel #3
0
    def _create_qresult(self, hit_blocks):
        """Create the Biopython data structures from the parsed data (PRIVATE)."""
        query_id = self.query_id
        hit_dict = OrderedDict()

        for output_index, block in enumerate(hit_blocks):
            hit_id = block['hit_id']

            frag = HSPFragment(hit_id, query_id)
            frag.alphabet = generic_protein
            frag.query_start = block['query_start'] - 1
            frag.query_end = block['query_end']
            frag.hit_start = block['hit_start'] - 1
            frag.hit_end = block['hit_end']
            frag.hit = block['hit_seq']
            frag.query = block['query_seq']

            hsp = HSP([frag])
            hsp.hit_id = hit_id
            hsp.output_index = output_index
            hsp.query_id = query_id
            hsp.hit_description = block['description']
            is_included = True  # Should everything should be included?
            hsp.is_included = is_included
            hsp.evalue = block['evalue']
            hsp.score = block['score']
            hsp.prob = block['prob']

            if hit_id not in hit_dict:
                hit = Hit([hsp], hit_id)
                hit.description = block['description']
                hit.is_included = is_included
                hit.evalue = block['evalue']
                hit.score = block['score']
                hit_dict[hit_id] = hit
            else:
                hit_dict[hit_id].append(hsp)

        qresult = QueryResult(hit_dict.values(), query_id)
        qresult.program = _PROGRAM
        qresult.seq_len = self.seq_len
        return [qresult]
Beispiel #4
0
def _create_hsp(hid, qid, psl):
    # protein flag
    is_protein = _is_protein(psl)
    # strand
    #if query is protein, strand is 0
    if is_protein:
        qstrand = 0
    else:
        qstrand = 1 if psl['strand'][0] == '+' else -1
    # try to get hit strand, if it exists
    try:
        hstrand = 1 if psl['strand'][1] == '+' else -1
    except IndexError:
        hstrand = 1  # hit strand defaults to plus

    # query block starts
    qstarts = _reorient_starts(psl['qstarts'], \
            psl['blocksizes'], psl['qsize'], qstrand)
    # hit block starts
    if len(psl['strand']) == 2:
        hstarts = _reorient_starts(psl['tstarts'], \
                psl['blocksizes'], psl['tsize'], hstrand)
    else:
        hstarts = psl['tstarts']
    # set query and hit coords
    # this assumes each block has no gaps (which seems to be the case)
    assert len(qstarts) == len(hstarts) == len(psl['blocksizes'])
    query_range_all = zip(qstarts, [x + y for x, y in \
            zip(qstarts, psl['blocksizes'])])
    hit_range_all = zip(hstarts, [x + y for x, y in \
            zip(hstarts, psl['blocksizes'])])
    # check length of sequences and coordinates, all must match
    if 'tseqs' in psl and 'qseqs' in psl:
        assert len(psl['tseqs']) == len(psl['qseqs']) == \
                len(query_range_all) == len(hit_range_all)
    else:
        assert len(query_range_all) == len(hit_range_all)

    frags = []
    # iterating over query_range_all, but hit_range_all works just as well
    for idx, qcoords in enumerate(query_range_all):
        hseqlist = psl.get('tseqs')
        hseq = '' if not hseqlist else hseqlist[idx]
        qseqlist = psl.get('qseqs')
        qseq = '' if not qseqlist else qseqlist[idx]
        frag = HSPFragment(hid, qid, hit=hseq, query=qseq)
        # set alphabet
        frag.alphabet = generic_dna
        # set coordinates
        frag.query_start = qcoords[0]
        frag.query_end = qcoords[1]
        frag.hit_start = hit_range_all[idx][0]
        frag.hit_end = hit_range_all[idx][1]
        # and strands
        frag.query_strand = qstrand
        frag.hit_strand = hstrand
        frags.append(frag)

    # create hsp object
    hsp = HSP(frags)
    # check if start and end are set correctly
    assert hsp.query_start == psl['qstart']
    assert hsp.query_end == psl['qend']
    assert hsp.hit_start == psl['tstart']
    assert hsp.hit_end == psl['tend']
    # and check block spans as well
    assert hsp.query_span_all == hsp.hit_span_all == psl['blocksizes']
    # set its attributes
    hsp.match_num = psl['matches']
    hsp.mismatch_num = psl['mismatches']
    hsp.match_rep_num = psl['repmatches']
    hsp.n_num = psl['ncount']
    hsp.query_gapopen_num = psl['qnuminsert']
    hsp.query_gap_num = psl['qbaseinsert']
    hsp.hit_gapopen_num = psl['tnuminsert']
    hsp.hit_gap_num = psl['tbaseinsert']

    hsp.ident_num = psl['matches'] + psl['repmatches']
    hsp.gapopen_num = psl['qnuminsert'] + psl['tnuminsert']
    hsp.gap_num = psl['qbaseinsert'] + psl['tbaseinsert']
    hsp.query_is_protein = is_protein
    hsp.ident_pct = 100.0 - _calc_millibad(psl, is_protein) * 0.1
    hsp.score = _calc_score(psl, is_protein)
    # helper flag, for writing
    hsp._has_hit_strand = len(psl['strand']) == 2

    return hsp
Beispiel #5
0
def _create_hsp(hid, qid, psl):
    """Create high scoring pair object (PRIVATE)."""
    # protein flag
    is_protein = _is_protein(psl)
    # strand
    # if query is protein, strand is 0
    if is_protein:
        qstrand = 0
    else:
        qstrand = 1 if psl["strand"][0] == "+" else -1
    # try to get hit strand, if it exists
    try:
        hstrand = 1 if psl["strand"][1] == "+" else -1
    except IndexError:
        hstrand = 1  # hit strand defaults to plus

    blocksize_multiplier = 3 if is_protein else 1
    # query block starts
    qstarts = _reorient_starts(psl["qstarts"], psl["blocksizes"], psl["qsize"],
                               qstrand)
    # hit block starts
    if len(psl["strand"]) == 2:
        hstarts = _reorient_starts(
            psl["tstarts"],
            [blocksize_multiplier * i for i in psl["blocksizes"]],
            psl["tsize"],
            hstrand,
        )
    else:
        hstarts = psl["tstarts"]
    # set query and hit coords
    # this assumes each block has no gaps (which seems to be the case)
    assert len(qstarts) == len(hstarts) == len(psl["blocksizes"])
    query_range_all = list(
        zip(qstarts, [x + y for x, y in zip(qstarts, psl["blocksizes"])]))
    hit_range_all = list(
        zip(
            hstarts,
            [
                x + y * blocksize_multiplier
                for x, y in zip(hstarts, psl["blocksizes"])
            ],
        ))
    # check length of sequences and coordinates, all must match
    if "tseqs" in psl and "qseqs" in psl:
        assert (len(psl["tseqs"]) == len(psl["qseqs"]) == len(query_range_all)
                == len(hit_range_all))
    else:
        assert len(query_range_all) == len(hit_range_all)

    frags = []
    # iterating over query_range_all, but hit_range_all works just as well
    for idx, qcoords in enumerate(query_range_all):
        hseqlist = psl.get("tseqs")
        hseq = "" if not hseqlist else hseqlist[idx]
        qseqlist = psl.get("qseqs")
        qseq = "" if not qseqlist else qseqlist[idx]
        frag = HSPFragment(hid, qid, hit=hseq, query=qseq)
        # set alphabet
        frag.alphabet = generic_dna
        # set coordinates
        frag.query_start = qcoords[0]
        frag.query_end = qcoords[1]
        frag.hit_start = hit_range_all[idx][0]
        frag.hit_end = hit_range_all[idx][1]
        # and strands
        frag.query_strand = qstrand
        frag.hit_strand = hstrand
        frags.append(frag)

    # create hsp object
    hsp = HSP(frags)
    # check if start and end are set correctly
    assert hsp.query_start == psl["qstart"]
    assert hsp.query_end == psl["qend"]
    assert hsp.hit_start == psl["tstart"]
    assert hsp.hit_end == psl["tend"]
    # and check block spans as well
    hit_spans = [span / blocksize_multiplier for span in hsp.hit_span_all]
    assert hit_spans == hsp.query_span_all == psl["blocksizes"]
    # set its attributes
    hsp.match_num = psl["matches"]
    hsp.mismatch_num = psl["mismatches"]
    hsp.match_rep_num = psl["repmatches"]
    hsp.n_num = psl["ncount"]
    hsp.query_gapopen_num = psl["qnuminsert"]
    hsp.query_gap_num = psl["qbaseinsert"]
    hsp.hit_gapopen_num = psl["tnuminsert"]
    hsp.hit_gap_num = psl["tbaseinsert"]

    hsp.ident_num = psl["matches"] + psl["repmatches"]
    hsp.gapopen_num = psl["qnuminsert"] + psl["tnuminsert"]
    hsp.gap_num = psl["qbaseinsert"] + psl["tbaseinsert"]
    hsp.query_is_protein = is_protein
    hsp.ident_pct = 100.0 - _calc_millibad(psl, is_protein) * 0.1
    hsp.score = _calc_score(psl, is_protein)
    # helper flag, for writing
    hsp._has_hit_strand = len(psl["strand"]) == 2

    return hsp
Beispiel #6
0
def _create_hsp(hid, qid, psl):
    # protein flag
    is_protein = _is_protein(psl)
    # strand
    # if query is protein, strand is 0
    if is_protein:
        qstrand = 0
    else:
        qstrand = 1 if psl['strand'][0] == '+' else -1
    # try to get hit strand, if it exists
    try:
        hstrand = 1 if psl['strand'][1] == '+' else -1
    except IndexError:
        hstrand = 1  # hit strand defaults to plus

    # query block starts
    qstarts = _reorient_starts(psl['qstarts'], psl['blocksizes'], psl['qsize'],
                               qstrand)
    # hit block starts
    if len(psl['strand']) == 2:
        hstarts = _reorient_starts(psl['tstarts'], psl['blocksizes'],
                                   psl['tsize'], hstrand)
    else:
        hstarts = psl['tstarts']
    # set query and hit coords
    # this assumes each block has no gaps (which seems to be the case)
    assert len(qstarts) == len(hstarts) == len(psl['blocksizes'])
    query_range_all = list(
        zip(qstarts, [x + y for x, y in zip(qstarts, psl['blocksizes'])]))
    hit_range_all = list(
        zip(hstarts, [x + y for x, y in zip(hstarts, psl['blocksizes'])]))
    # check length of sequences and coordinates, all must match
    if 'tseqs' in psl and 'qseqs' in psl:
        assert len(psl['tseqs']) == len(psl['qseqs']) == \
                len(query_range_all) == len(hit_range_all)
    else:
        assert len(query_range_all) == len(hit_range_all)

    frags = []
    # iterating over query_range_all, but hit_range_all works just as well
    for idx, qcoords in enumerate(query_range_all):
        hseqlist = psl.get('tseqs')
        hseq = '' if not hseqlist else hseqlist[idx]
        qseqlist = psl.get('qseqs')
        qseq = '' if not qseqlist else qseqlist[idx]
        frag = HSPFragment(hid, qid, hit=hseq, query=qseq)
        # set alphabet
        frag.alphabet = generic_dna
        # set coordinates
        frag.query_start = qcoords[0]
        frag.query_end = qcoords[1]
        frag.hit_start = hit_range_all[idx][0]
        frag.hit_end = hit_range_all[idx][1]
        # and strands
        frag.query_strand = qstrand
        frag.hit_strand = hstrand
        frags.append(frag)

    # create hsp object
    hsp = HSP(frags)
    # check if start and end are set correctly
    assert hsp.query_start == psl['qstart']
    assert hsp.query_end == psl['qend']
    assert hsp.hit_start == psl['tstart']
    assert hsp.hit_end == psl['tend']
    # and check block spans as well
    assert hsp.query_span_all == hsp.hit_span_all == psl['blocksizes']
    # set its attributes
    hsp.match_num = psl['matches']
    hsp.mismatch_num = psl['mismatches']
    hsp.match_rep_num = psl['repmatches']
    hsp.n_num = psl['ncount']
    hsp.query_gapopen_num = psl['qnuminsert']
    hsp.query_gap_num = psl['qbaseinsert']
    hsp.hit_gapopen_num = psl['tnuminsert']
    hsp.hit_gap_num = psl['tbaseinsert']

    hsp.ident_num = psl['matches'] + psl['repmatches']
    hsp.gapopen_num = psl['qnuminsert'] + psl['tnuminsert']
    hsp.gap_num = psl['qbaseinsert'] + psl['tbaseinsert']
    hsp.query_is_protein = is_protein
    hsp.ident_pct = 100.0 - _calc_millibad(psl, is_protein) * 0.1
    hsp.score = _calc_score(psl, is_protein)
    # helper flag, for writing
    hsp._has_hit_strand = len(psl['strand']) == 2

    return hsp