def test_subpoagraph_construction_full_graph(self): nodes = [ graph.Node(node_id=nid(0), base=b('A'), aligned_to=None), graph.Node(node_id=nid(1), base=b('A'), aligned_to=None), graph.Node(node_id=nid(2), base=b('C'), aligned_to=None), graph.Node(node_id=nid(3), base=b('A'), aligned_to=None), graph.Node(node_id=nid(4), base=b('T'), aligned_to=None) ] sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 4])])], graph.SequenceMetadata({'group': '1'})) } poagraph = graph.Poagraph(nodes, sequences) translator = poa._PoagraphPOTranslator(poagraph, [msa.SequenceID('seq0')]) actual_po_content = translator.get_input_po_content() expected_po_content = "VERSION=pangenome\n" \ "NAME=pangenome\n" \ "TITLE=pangenome\n" \ "LENGTH=5\n" \ "SOURCECOUNT=1\n" \ "SOURCENAME=seq0\n" \ "SOURCEINFO=5 0 100 -1 seq0\n" \ "a:S0\n" \ "a:L0S0\n" \ "c:L1S0\n" \ "a:L2S0\n" \ "t:L3S0" self.assertEqual(expected_po_content, actual_po_content)
def test_subpoagraph_should_omit_edges_2(self): nodes = [ graph.Node(node_id=nid(0), base=b('A'), aligned_to=None), graph.Node(node_id=nid(1), base=b('C'), aligned_to=None), graph.Node(node_id=nid(2), base=b('C'), aligned_to=None) ] sequences = { msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 2])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 1, 2])])], graph.SequenceMetadata({'group': '1'})) } poagraph = graph.Poagraph(nodes, sequences) translator = poa._PoagraphPOTranslator(poagraph, [msa.SequenceID('seq1')]) actual_po_content = translator.get_input_po_content() expected_po_content = "VERSION=pangenome\n" \ "NAME=pangenome\n" \ "TITLE=pangenome\n" \ "LENGTH=2\n" \ "SOURCECOUNT=1\n" \ "SOURCENAME=seq1\n" \ "SOURCEINFO=2 0 100 -1 seq1\n" \ "a:S0\n" \ "c:L0S0" self.assertEqual(expected_po_content, actual_po_content)
def test_1_p_parameter_influence(self, p: at_params.P, expected_cutoff: graph.Compatibility): nodes = [ graph.Node(node_id=nid(0), base=b('T'), aligned_to=None), graph.Node(node_id=nid(1), base=b('A'), aligned_to=None), graph.Node(node_id=nid(2), base=b('G'), aligned_to=None), graph.Node(node_id=nid(3), base=b('A'), aligned_to=None), graph.Node(node_id=nid(4), base=b('C'), aligned_to=None), graph.Node(node_id=nid(5), base=b('A'), aligned_to=None), graph.Node(node_id=nid(6), base=b('C'), aligned_to=None), graph.Node(node_id=nid(7), base=b('G'), aligned_to=None), graph.Node(node_id=nid(8), base=b('T'), aligned_to=None), graph.Node(node_id=nid(9), base=b('A'), aligned_to=None) ] sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [ graph.SeqPath( [*map(nid, [10, 11, 12, 13, 14, 15, 16, 17, 18, 9])]) ], graph.SequenceMetadata({})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [ graph.SeqPath( [*map(nid, [10, 11, 12, 13, 14, 15, 16, 17, 8, 9])]) ], graph.SequenceMetadata({})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [ graph.SeqPath( [*map(nid, [10, 11, 12, 13, 14, 15, 16, 7, 8, 9])]) ], graph.SequenceMetadata({})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [ graph.SeqPath([*map(nid, [10, 11, 12, 3, 4, 5, 6, 7, 8, 9])]) ], graph.SequenceMetadata({})), msa.SequenceID('seq4'): graph.Sequence( msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [10, 11, 2, 3, 4, 5, 6, 7, 8, 9])])], graph.SequenceMetadata({})) } poagraph = graph.Poagraph(nodes, sequences) consensus_path = graph.SeqPath( [*map(nid, [10, 11, 12, 13, 14, 15, 16, 17, 18, 19])]) compatibilities = poagraph.get_compatibilities( poagraph.get_sequences_ids(), consensus_path, p) actual_cutoff = at_builders._find_node_cutoff( [c for c in compatibilities.values()], []).cutoff self.assertAlmostEqual(expected_cutoff.value, actual_cutoff.value)
def test_subpoagraph_unfilled_nodes(self): symbol_for_uknown = '?' nodes = [ graph.Node(node_id=nid(0), base=b('A'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=b('C'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=b('G'), aligned_to=None), graph.Node(node_id=nid(3), base=b(symbol_for_uknown), aligned_to=None), graph.Node(node_id=nid(4), base=b(symbol_for_uknown), aligned_to=None), graph.Node(node_id=nid(5), base=b('G'), aligned_to=None), graph.Node(node_id=nid(6), base=b('C'), aligned_to=None), graph.Node(node_id=nid(7), base=b('A'), aligned_to=None), graph.Node(node_id=nid(5), base=b('T'), aligned_to=None) ] sequences = { msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 2, 3, 4, 7, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [1, 2, 5, 6, 7, 8])])], graph.SequenceMetadata({'group': '1'})) } poagraph = graph.Poagraph(nodes, sequences) translator = poa._PoagraphPOTranslator( poagraph, [msa.SequenceID('seq1'), msa.SequenceID('seq2')]) actual_po_content = translator.get_input_po_content() expected_po_content = "VERSION=pangenome\n" \ "NAME=pangenome\n" \ "TITLE=pangenome\n" \ "LENGTH=9\n" \ "SOURCECOUNT=2\n" \ "SOURCENAME=seq1\n" \ "SOURCEINFO=6 0 100 -1 seq1\n" \ "SOURCENAME=seq2\n" \ "SOURCEINFO=6 1 100 -1 seq2\n" \ "a:S0A1\n" \ "c:S1A0\n" \ "g:L0L1S0S1\n" \ f"{symbol_for_uknown}:L2S0\n" \ f"{symbol_for_uknown}:L3S0\n" \ "g:L2S1\n" \ "c:L5S1\n" \ "a:L4L6S0S1\n" \ "t:L7S0S1" self.assertEqual(expected_po_content, actual_po_content)
def test_05_single_block_single_nucletodide(self): maf_path = self.maf_files_dir.joinpath( "test_5_single_block_single_nucletodide.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=None, block_id=bid(0)) ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [0])])], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def _complement_sequence_starting_nodes(build_state: _BuildState, seq_id: msa.SequenceID, first_block_sinfo: SequenceInfo) -> \ None: current_node_id: graph.NodeID = _get_max_node_id(build_state.nodes) column_id = -first_block_sinfo.start join_with = None for i in range(first_block_sinfo.start): current_node_id += 1 missing_nucleotide = _get_missing_nucleotide( build_state.fasta_provider, seq_id, i) build_state.nodes += [ graph.Node(node_id=current_node_id, base=missing_nucleotide, column_id=column_id) ] _add_node_to_sequence(build_state, seq_id=seq_id, join_with=join_with, node_id=current_node_id) join_with = current_node_id column_id += 1 build_state.free_edges[seq_id] += [ Edge(seq_id=seq_id, from_block_id=None, to_block_id=first_block_sinfo.block_id, last_node_id=current_node_id) ]
def _complement_sequence_middle_nodes(build_state: _BuildState, seq_id: msa.SequenceID, last_pos, next_pos, last_node_id: graph.NodeID) -> \ graph.NodeID: current_node_id = _get_max_node_id(build_state.nodes) column_id = build_state.column_id join_with = last_node_id for i in range(last_pos + 1, next_pos): column_id += 1 current_node_id += 1 missing_nucleotide = _get_missing_nucleotide( build_state.fasta_provider, seq_id, i) build_state.nodes += [ graph.Node(node_id=current_node_id, base=missing_nucleotide, aligned_to=None, column_id=column_id, block_id=None) ] _add_node_to_sequence(build_state, seq_id=seq_id, join_with=join_with, node_id=current_node_id) join_with = current_node_id return current_node_id
def get_poagraph(maf: msa.Maf, metadata: Optional[msa.MetadataCSV]) -> \ Tuple[List[graph.Node], Dict[msa.SequenceID, graph.Sequence]]: """Get poagraph elements from MAF. Args: maf: Multialignment file in MAF format. metadata: MetadataCSV. Returns: Tuple of poagraph elements. """ alignment = [*AlignIO.parse(maf.filecontent, "maf")] nodes, sequences = _init_poagraph(alignment, metadata) current_node_id = graph.NodeID(-1) column_id = graph.ColumnID(-1) for block_id, block in enumerate(alignment): global_logger.info(f"Processing block {block_id}...") block_width = len(block[0].seq) for col in range(block_width): column_id += 1 sequence_id_to_nucleotide = { msa.SequenceID(seq.id): seq[col] for seq in block } nodes_codes = sorted([ *(set([ nucleotide for nucleotide in sequence_id_to_nucleotide.values() ])).difference({'-'}) ]) column_nodes_ids = [ graph.NodeID(current_node_id + i + 1) for i, _ in enumerate(nodes_codes) ] for i, nucl in enumerate(nodes_codes): current_node_id += 1 nodes.append( graph.Node(node_id=current_node_id, base=graph.Base(nucl), aligned_to=_get_next_aligned_node_id( graph.NodeID(i), column_nodes_ids), column_id=graph.ColumnID(column_id), block_id=graph.BlockID(block_id))) for seq_id, nucleotide in sequence_id_to_nucleotide.items(): if nucleotide == nucl: sequences[seq_id] = _add_node_do_sequence( sequence=sequences[seq_id], node_id=current_node_id) return nodes, sequences
def test_08_reversed_block(self): maf_path = self.maf_files_dir.joinpath("test_8_reversed_block.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(1), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(2), base=graph.Base('T'), aligned_to=None), # next block is reversed because it was converted to dag graph.Node(node_id=nid(3), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(4), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(5), base=graph.Base('A'), aligned_to=nid(6)), graph.Node(node_id=nid(6), base=graph.Base('G'), aligned_to=nid(5)), graph.Node(node_id=nid(7), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(8), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(9), base=graph.Base('T'), aligned_to=None), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 1, 3, 4, 5, 7, 8, 9])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 4, 6, 7, 8, 9])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 4, 6, 7, 9])])], graph.SequenceMetadata({'group': '2'})), } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_2_consensuses_and_empty_sequences(self): expected_po_content_path = self.po_files_dir.joinpath("test_2.po") poagraph_nodes = [graph.Node(node_id=nid(0), base=bid('C'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=bid('T'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=bid('A'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=bid('G'), aligned_to=nid(2)), graph.Node(node_id=nid(4), base=bid('C'), aligned_to=None), graph.Node(node_id=nid(5), base=bid('T'), aligned_to=None), graph.Node(node_id=nid(6), base=bid('A'), aligned_to=nid(7)), graph.Node(node_id=nid(7), base=bid('T'), aligned_to=nid(6)), graph.Node(node_id=nid(8), base=bid('G'), aligned_to=None) ] poagraph_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 3, 4, 5, 6, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 2, 4, 5, 7, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('CONSENS0'): graph.Sequence(msa.SequenceID('CONSENS0'), [graph.SeqPath([*map(nid, [0, 3, 4, 5, 7, 8])])], None), msa.SequenceID('CONSENS1'): graph.Sequence(msa.SequenceID('CONSENS1'), [graph.SeqPath([*map(nid, [1, 2, 4, 5, 6, 8])])], None), } poagraph = graph.Poagraph(poagraph_nodes, poagraph_sequences) actual_po_content = po.poagraph_to_PangenomePO(poagraph) expected_po_content = pathtools.get_file_content(expected_po_content_path) self.assertEqual(expected_po_content, actual_po_content)
def test_2_consensuses_and_empty_sequences(self): po_path = self.po_files_dir.joinpath("test_2.po") expected_nodes = [ graph.Node(node_id=nid(0), base=bid('C'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=bid('T'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=bid('A'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=bid('G'), aligned_to=nid(2)), graph.Node(node_id=nid(4), base=bid('C'), aligned_to=None), graph.Node(node_id=nid(5), base=bid('T'), aligned_to=None), graph.Node(node_id=nid(6), base=bid('A'), aligned_to=nid(7)), graph.Node(node_id=nid(7), base=bid('T'), aligned_to=nid(6)), graph.Node(node_id=nid(8), base=bid('G'), aligned_to=None) ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 3, 4, 5, 6, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 2, 4, 5, 7, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('CONSENS0'): graph.Sequence(msa.SequenceID('CONSENS0'), [graph.SeqPath([*map(nid, [0, 3, 4, 5, 7, 8])])], graph.SequenceMetadata({})), msa.SequenceID('CONSENS1'): graph.Sequence(msa.SequenceID('CONSENS1'), [graph.SeqPath([*map(nid, [1, 2, 4, 5, 6, 8])])], graph.SequenceMetadata({})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) nodes, sequences = po2poagraph.get_poagraph( msa.Po(pathtools.get_file_content_stringio(po_path), po_path), self.metadatacsv) actual_poagraph = graph.Poagraph(nodes, sequences) self.assertEqual(expected_poagraph, actual_poagraph)
def test_02_seq_starts_in_second_block(self): maf_path = self.maf_files_dir.joinpath( "test_2_seq_starts_in_second_block.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('C'), aligned_to=None, block_id=bid(0)), graph.Node(node_id=nid(1), base=graph.Base('T'), aligned_to=None, block_id=bid(0)), graph.Node(node_id=nid(2), base=graph.Base('G'), aligned_to=None, block_id=bid(0)), graph.Node(node_id=nid(3), base=graph.Base('T'), aligned_to=None, block_id=bid(1)), graph.Node(node_id=nid(4), base=graph.Base('G'), aligned_to=nid(5), block_id=bid(2)), graph.Node(node_id=nid(5), base=graph.Base('T'), aligned_to=nid(4), block_id=bid(2)), graph.Node(node_id=nid(6), base=graph.Base('A'), aligned_to=None, block_id=bid(2)), graph.Node(node_id=nid(7), base=graph.Base('A'), aligned_to=None, block_id=bid(2)), graph.Node(node_id=nid(8), base=graph.Base('C'), aligned_to=None, block_id=bid(2)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [1, 2, 3])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 5, 7])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [3, 4, 6, 8])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def _complement_sequence_middles_if_needed(build_state: _BuildState, block: Block, edge: Arc, seq, last_node_id: graph.NodeID): seq_id = msa.SequenceID(seq[0].seq_id) left_block_sinfo, right_block_sinfo = _get_edge_sinfos( seqs_info=build_state.seqs_info, from_block_id=block.id, edge=edge, seq_id=seq_id) if _complementation_not_needed(left_block_sinfo, right_block_sinfo): if edge.edge_type == (1, -1): return last_node_id else: return None else: current_node_id = _get_max_node_id(build_state.nodes) column_id = build_state.column_id if left_block_sinfo.start < right_block_sinfo.start: last_pos = left_block_sinfo.start + left_block_sinfo.size - 1 next_pos = right_block_sinfo.start else: last_pos = right_block_sinfo.start + right_block_sinfo.size - 1 next_pos = left_block_sinfo.start join_with = last_node_id if _should_join_with_last_node( edge.edge_type) else None for i in range(last_pos + 1, next_pos): column_id += 1 current_node_id += 1 missing_nucleotide = _get_missing_nucleotide( build_state.fasta_provider, seq_id, i) build_state.nodes += [ graph.Node(node_id=current_node_id, base=missing_nucleotide, aligned_to=None, column_id=column_id, block_id=None) ] _add_node_to_sequence(build_state, seq_id=seq_id, join_with=join_with, node_id=current_node_id) join_with = current_node_id if _should_join_with_next_node(edge.edge_type): return current_node_id else: return None
def _process_block(build_state: _BuildState, block: DAGMaf.DAGMafNode): current_node_id = _get_max_node_id(build_state.nodes) block_width = len(block.alignment[0].seq) paths_join_info = _get_paths_join_info(block, build_state.free_edges) build_state.column_id = _get_max_column_id(build_state.nodes) for col in range(block_width): build_state.column_id += 1 sequence_name_to_nucleotide = { MafSequenceID(seq.id): seq[col] for seq in block.alignment } nodes_codes = _get_column_nucleotides_sorted_codes( sequence_name_to_nucleotide) column_nodes_ids = [ current_node_id + i + 1 for i, _ in enumerate(nodes_codes) ] for i, nucl in enumerate(nodes_codes): current_node_id += 1 maf_seqs_id = [ seq_id for seq_id, n in sequence_name_to_nucleotide.items() if n == nucl ] build_state.nodes += [ graph.Node(node_id=current_node_id, base=graph.Base(nucl), aligned_to=_get_next_aligned_node_id( i, column_nodes_ids), column_id=build_state.column_id, block_id=block.id) ] for maf_seq_id in maf_seqs_id: seq_id = msa.SequenceID(maf_seq_id) _add_node_to_sequence(build_state, seq_id, paths_join_info[seq_id], current_node_id) paths_join_info[seq_id] = current_node_id _add_block_out_edges_to_free_edges(build_state, block, paths_join_info) _manage_endings(build_state, block, paths_join_info)
def _get_poagraph_paths_and_nodes(po_lines: List[str], sequences_info: Dict[int, POSequenceInfo], sequences: Dict[msa.SequenceID, graph.Sequence]) -> \ Tuple[List[graph.Node], Dict[msa.SequenceID, graph.Sequence]]: nodes_count = int(_extract_line_value(po_lines[3])) paths_count = int(_extract_line_value(po_lines[4])) nodes: List[graph.Node] = [None] * nodes_count node_id = 0 for i in range(5 + paths_count * 2, 5 + paths_count * 2 + nodes_count): node_line = po_lines[i] base = graph.Base(node_line[0].upper()) in_nodes, po_sequences_ids, aligned_to = _extract_node_parameters(node_line) sequences_ids = [sequences_info[po_sequences_id].name for po_sequences_id in po_sequences_ids] nodes[node_id] = graph.Node(graph.NodeID(node_id), base, graph.NodeID(aligned_to)) for seq_id in sequences_ids: if len(sequences[seq_id].paths) == 1: sequences[seq_id].paths[0].append(graph.NodeID(node_id)) else: sequences[seq_id].paths.append(graph.SeqPath([graph.NodeID(node_id)])) node_id += 1 return nodes, sequences
def test_1_typical_poagraph(self): po_path = self.po_files_dir.joinpath("test_1.po") expected_nodes = [ graph.Node(node_id=nid(0), base=bid('A'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=bid('G'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=bid('C'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=bid('G'), aligned_to=nid(2)), graph.Node(node_id=nid(4), base=bid('A'), aligned_to=nid(5)), graph.Node(node_id=nid(5), base=bid('T'), aligned_to=nid(4)), graph.Node(node_id=nid(6), base=bid('G'), aligned_to=None), graph.Node(node_id=nid(7), base=bid('G'), aligned_to=None), graph.Node(node_id=nid(8), base=bid('A'), aligned_to=nid(9)), graph.Node(node_id=nid(9), base=bid('C'), aligned_to=nid(10)), graph.Node(node_id=nid(10), base=bid('G'), aligned_to=nid(11)), graph.Node(node_id=nid(11), base=bid('T'), aligned_to=nid(8)), graph.Node(node_id=nid(12), base=bid('A'), aligned_to=nid(13)), graph.Node(node_id=nid(13), base=bid('C'), aligned_to=nid(12)), graph.Node(node_id=nid(14), base=bid('T'), aligned_to=None), graph.Node(node_id=nid(15), base=bid('A'), aligned_to=nid(16)), graph.Node(node_id=nid(16), base=bid('C'), aligned_to=nid(17)), graph.Node(node_id=nid(17), base=bid('G'), aligned_to=nid(15)) ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence( msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 2, 4, 6, 7, 8, 12, 14, 16])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 2, 5, 6, 7, 9])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence( msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [3, 4, 6, 7, 10, 12, 14, 17])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [11, 13, 14, 15])])], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) nodes, sequences = po2poagraph.get_poagraph( msa.Po(pathtools.get_file_content_stringio(po_path), po_path), self.metadatacsv) actual_poagraph = graph.Poagraph(nodes, sequences) self.assertEqual(expected_poagraph, actual_poagraph)
def test_1_typical_poagraph(self): expected_po_content_path = self.po_files_dir.joinpath("test_1.po") poagraph_nodes = [graph.Node(node_id=nid(0), base=bid('A'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=bid('G'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=bid('C'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=bid('G'), aligned_to=nid(2)), graph.Node(node_id=nid(4), base=bid('A'), aligned_to=nid(5)), graph.Node(node_id=nid(5), base=bid('T'), aligned_to=nid(4)), graph.Node(node_id=nid(6), base=bid('G'), aligned_to=None), graph.Node(node_id=nid(7), base=bid('G'), aligned_to=None), graph.Node(node_id=nid(8), base=bid('A'), aligned_to=nid(9)), graph.Node(node_id=nid(9), base=bid('C'), aligned_to=nid(10)), graph.Node(node_id=nid(10), base=bid('G'), aligned_to=nid(11)), graph.Node(node_id=nid(11), base=bid('T'), aligned_to=nid(8)), graph.Node(node_id=nid(12), base=bid('A'), aligned_to=nid(13)), graph.Node(node_id=nid(13), base=bid('C'), aligned_to=nid(12)), graph.Node(node_id=nid(14), base=bid('T'), aligned_to=None), graph.Node(node_id=nid(15), base=bid('A'), aligned_to=nid(16)), graph.Node(node_id=nid(16), base=bid('C'), aligned_to=nid(17)), graph.Node(node_id=nid(17), base=bid('G'), aligned_to=nid(15)) ] poagraph_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 2, 4, 6, 7, 8, 12, 14, 16])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 2, 5, 6, 7, 9])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [3, 4, 6, 7, 10, 12, 14, 17])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [11, 13, 14, 15])])], graph.SequenceMetadata({'group': '1'})), } poagraph = graph.Poagraph(poagraph_nodes, poagraph_sequences) actual_po_content = po.poagraph_to_PangenomePO(poagraph) expected_po_content = pathtools.get_file_content(expected_po_content_path) self.assertEqual(expected_po_content, actual_po_content)
def setUp(self): nodes = [ graph.Node( node_id=nid(0), base=b('T'), aligned_to=None, ), graph.Node(node_id=nid(1), base=b('A'), aligned_to=nid(2)), graph.Node(node_id=nid(2), base=b('G'), aligned_to=nid(1)), graph.Node(node_id=nid(3), base=b('A'), aligned_to=nid(4)), graph.Node(node_id=nid(4), base=b('C'), aligned_to=nid(3)), graph.Node(node_id=nid(5), base=b('A'), aligned_to=nid(6)), graph.Node(node_id=nid(6), base=b('C'), aligned_to=nid(7)), graph.Node(node_id=nid(7), base=b('G'), aligned_to=nid(8)), graph.Node(node_id=nid(8), base=b('T'), aligned_to=nid(5)), graph.Node(node_id=nid(9), base=b('A'), aligned_to=None), graph.Node(node_id=nid(10), base=b('C'), aligned_to=nid(11)), graph.Node(node_id=nid(11), base=b('T'), aligned_to=nid(10)), graph.Node(node_id=nid(12), base=b('G'), aligned_to=None), graph.Node(node_id=nid(13), base=b('A'), aligned_to=nid(14)), graph.Node(node_id=nid(14), base=b('C'), aligned_to=nid(13)) ] sequences = { msa.SequenceID('seq0'): graph.Sequence( msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 1, 3, 5, 9, 10, 13])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 3, 6, 9, 11])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [2, 4, 7, 9, 11, 12])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq3'): graph.Sequence( msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [2, 4, 8, 9, 11, 12, 14])])], graph.SequenceMetadata({'group': '1'})), } self.poagraph = graph.Poagraph(nodes, sequences)
def test_09_inactive_edges_but_all_strands_plus(self): maf_path = self.maf_files_dir.joinpath("test_9_inactive_edges_but_all_strands_plus.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(1), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(2), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(3), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(4), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(5), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(6), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(7), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(8), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(9), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(10), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(11), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(12), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(13), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(14), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(15), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(16), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(17), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(18), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(19), base=graph.Base('G'), aligned_to=None), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 4, 10, 11, 12, 13, 14])]), graph.SeqPath([*map(nid, [5, 6, 7, 8, 9, 15, 16, 17, 18, 19])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})), } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_10_parallel_blocks_1st_and_2nd_merge_into_3rd(self): maf_path = self.maf_files_dir.joinpath("test_10_parallel_blocks_1st_and_2nd_merge_into_3rd.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('G'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=graph.Base('T'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(3), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(4), base=graph.Base('C'), aligned_to=nid(5)), graph.Node(node_id=nid(5), base=graph.Base('G'), aligned_to=nid(4)), graph.Node(node_id=nid(6), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(7), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(8), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(9), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(10), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(11), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(12), base=graph.Base('C'), aligned_to=nid(13)), graph.Node(node_id=nid(13), base=graph.Base('G'), aligned_to=nid(12)), graph.Node(node_id=nid(14), base=graph.Base('C'), aligned_to=nid(15)), graph.Node(node_id=nid(15), base=graph.Base('G'), aligned_to=nid(16)), graph.Node(node_id=nid(16), base=graph.Base('T'), aligned_to=nid(14)), graph.Node(node_id=nid(17), base=graph.Base('A'), aligned_to=nid(18)), graph.Node(node_id=nid(18), base=graph.Base('T'), aligned_to=nid(17)), graph.Node(node_id=nid(19), base=graph.Base('A'), aligned_to=nid(20)), graph.Node(node_id=nid(20), base=graph.Base('C'), aligned_to=nid(19)), graph.Node(node_id=nid(21), base=graph.Base('C'), aligned_to=nid(22)), graph.Node(node_id=nid(22), base=graph.Base('G'), aligned_to=nid(21)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [7, 8, 9, 10, 11, 12, 15, 18, 19, 21])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [7, 8, 9, 10, 11, 12, 15, 18, 19, 21])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 2, 3, 4, 6, 13, 16, 17, 20, 21])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [graph.SeqPath([*map(nid, [1, 2, 3, 5, 6, 13, 14, 17, 20, 22])])], graph.SequenceMetadata({'group': '2'})), } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_1_messy_sequences(self): maf_path = self.maf_files_dir.joinpath("test_1_messy_sequences.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=None, block_id=bid(0)), graph.Node(node_id=nid(1), base=graph.Base('A'), aligned_to=nid(2), block_id=bid(0)), graph.Node(node_id=nid(2), base=graph.Base('C'), aligned_to=nid(1), block_id=bid(0)), graph.Node(node_id=nid(3), base=graph.Base('T'), aligned_to=None, block_id=bid(0)), graph.Node(node_id=nid(4), base=graph.Base('C'), aligned_to=nid(5), block_id=bid(0)), graph.Node(node_id=nid(5), base=graph.Base('G'), aligned_to=nid(4), block_id=bid(0)), graph.Node(node_id=nid(6), base=graph.Base('A'), aligned_to=None, block_id=bid(1)), graph.Node(node_id=nid(7), base=graph.Base('C'), aligned_to=None, block_id=bid(1)), graph.Node(node_id=nid(8), base=graph.Base('G'), aligned_to=None, block_id=bid(1)), graph.Node(node_id=nid(9), base=graph.Base('C'), aligned_to=nid(10), block_id=bid(2)), graph.Node(node_id=nid(10), base=graph.Base('G'), aligned_to=nid(9), block_id=bid(2)), graph.Node(node_id=nid(11), base=graph.Base('T'), aligned_to=None, block_id=bid(2)), graph.Node(node_id=nid(12), base=graph.Base('C'), aligned_to=None, block_id=bid(2)), graph.Node(node_id=nid(13), base=graph.Base('C'), aligned_to=None, block_id=bid(2)), graph.Node(node_id=nid(14), base=graph.Base('A'), aligned_to=None, block_id=bid(2)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence( msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [1, 3, 4, 6, 8, 9, 11, 12])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence( msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [2, 3, 4, 10, 11, 12, 13, 14])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence( msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 2, 5, 6, 7, 10, 11, 12, 14])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } actual_nodes, actual_sequences = maf2poagraph.get_poagraph( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.metadatacsv) self.assertEqual(expected_nodes, actual_nodes) self.assertEqual(expected_sequences, actual_sequences)
def test_06_1st_block_separates_into_2_branches_which_connect_in_3rd_block(self): maf_path = self.maf_files_dir.joinpath( "test_6_1st_block_separates_into_2_branches_which_connect_in_3rd_block.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=graph.Base('C'), aligned_to=nid(2)), graph.Node(node_id=nid(2), base=graph.Base('G'), aligned_to=nid(0)), graph.Node(node_id=nid(3), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(4), base=graph.Base('A'), aligned_to=nid(5)), graph.Node(node_id=nid(5), base=graph.Base('T'), aligned_to=nid(4)), graph.Node(node_id=nid(6), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(7), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(8), base=graph.Base('C'), aligned_to=nid(9)), graph.Node(node_id=nid(9), base=graph.Base('G'), aligned_to=nid(10)), graph.Node(node_id=nid(10), base=graph.Base('T'), aligned_to=nid(8)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [graph.SeqPath([*map(nid, [0, 3, 4, 8])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [1, 3, 5, 6, 7, 9])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence(msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [2, 3, 5, 10])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_7_missing_one_reverted_sequence_middle_minus1_minus1(self): maf_path = self.maf_files_dir.joinpath( "test_7_missing_one_reverted_sequence_middle_minus1_minus1.maf") expected_nodes = [ # block 0 graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(1), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(2), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(3), base=graph.Base('A'), aligned_to=None), # missing seq2 graph.Node(node_id=nid(4), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(5), base=graph.Base('A'), aligned_to=None), # block 1 graph.Node(node_id=nid(6), base=graph.Base('A'), aligned_to=nid(7)), graph.Node(node_id=nid(7), base=graph.Base('G'), aligned_to=nid(6)), graph.Node(node_id=nid(8), base=graph.Base('C'), aligned_to=nid(9)), graph.Node(node_id=nid(9), base=graph.Base('G'), aligned_to=nid(8)), graph.Node(node_id=nid(10), base=graph.Base('C'), aligned_to=nid(11)), graph.Node(node_id=nid(11), base=graph.Base('T'), aligned_to=nid(10)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence( msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 1, 2, 3, 7, 9, 11])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence( msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 1, 4, 5, 6, 8, 10])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), self.fasta_provider, self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_6_missing_one_reverted_sequence_middle_minus1_1(self): maf_path = self.maf_files_dir.joinpath( "test_6_missing_one_reverted_sequence_middle_minus1_1.maf") expected_nodes = [ # block 1 because it is first in DAG and reverted graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(1), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(2), base=graph.Base('C'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=graph.Base('T'), aligned_to=nid(2)), # missing seq2, on edge (-1,1) graph.Node(node_id=nid(4), base=graph.Base(self.missing_n.value), aligned_to=None), graph.Node(node_id=nid(5), base=graph.Base(self.missing_n.value), aligned_to=None), graph.Node(node_id=nid(6), base=graph.Base('A'), aligned_to=nid(7)), graph.Node(node_id=nid(7), base=graph.Base('C'), aligned_to=nid(6)), graph.Node(node_id=nid(8), base=graph.Base('C'), aligned_to=None), graph.Node(node_id=nid(9), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(10), base=graph.Base('A'), aligned_to=nid(11)), graph.Node(node_id=nid(11), base=graph.Base('C'), aligned_to=nid(10)), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence(msa.SequenceID('seq1'), [ graph.SeqPath([*map(nid, [0, 1, 2])]), graph.SeqPath([*map(nid, [6, 8, 9, 10])]) ], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence( msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [0, 1, 3, 4, 5, 7, 11])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), missings.ConstBaseProvider(self.missing_n), self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)
def test_2_missing_sequence_end(self): maf_path = self.maf_files_dir.joinpath( "test_2_missing_sequence_end.maf") expected_nodes = [ graph.Node(node_id=nid(0), base=graph.Base('A'), aligned_to=nid(1)), graph.Node(node_id=nid(1), base=graph.Base('G'), aligned_to=nid(0)), graph.Node(node_id=nid(2), base=graph.Base('C'), aligned_to=nid(3)), graph.Node(node_id=nid(3), base=graph.Base('G'), aligned_to=nid(2)), graph.Node(node_id=nid(4), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(5), base=graph.Base('A'), aligned_to=nid(6)), graph.Node(node_id=nid(6), base=graph.Base('C'), aligned_to=nid(5)), graph.Node(node_id=nid(7), base=graph.Base('A'), aligned_to=None), graph.Node(node_id=nid(8), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(9), base=graph.Base('G'), aligned_to=None), graph.Node(node_id=nid(10), base=graph.Base('T'), aligned_to=None), graph.Node(node_id=nid(11), base=graph.Base(self.missing_n.value), aligned_to=None), graph.Node(node_id=nid(12), base=graph.Base(self.missing_n.value), aligned_to=None), ] expected_sequences = { msa.SequenceID('seq0'): graph.Sequence(msa.SequenceID('seq0'), [], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq1'): graph.Sequence( msa.SequenceID('seq1'), [graph.SeqPath([*map(nid, [0, 2, 4, 5, 8, 9, 10])])], graph.SequenceMetadata({'group': '1'})), msa.SequenceID('seq2'): graph.Sequence( msa.SequenceID('seq2'), [graph.SeqPath([*map(nid, [1, 3, 4, 6, 7, 11, 12])])], graph.SequenceMetadata({'group': '2'})), msa.SequenceID('seq3'): graph.Sequence(msa.SequenceID('seq3'), [], graph.SequenceMetadata({'group': '2'})) } expected_poagraph = graph.Poagraph(expected_nodes, expected_sequences) actual_poagraph, _ = builder.build_from_dagmaf( msa.Maf(pathtools.get_file_content_stringio(maf_path), maf_path), missings.ConstBaseProvider(self.missing_n), self.metadatacsv) self.assertEqual(expected_poagraph, actual_poagraph)