def test_1_no_symbol_provided(self):
missing_symbol = missings.MissingBase()
const_symbol_provider = missings.ConstBaseProvider(missing_symbol)
expected_symbol = graph.Base('?')
actual_symbol = const_symbol_provider.get_base(msa.SequenceID('s'), 0)
self.assertEqual(expected_symbol, actual_symbol)
def raise_error_if_unequal(self, sequence_id: msa.SequenceID,
expected_sequence: str,
fasta_provider: missings.FromFile) -> None:
for i, expected_symbol in enumerate(expected_sequence):
expected_base = graph.Base(expected_symbol)
actual_base = fasta_provider.get_base(sequence_id, i)
self.assertEqual(expected_base, actual_base)
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 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_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 get_base(self, seq_id: msa.SequenceID, i: int) -> graph.Base:
"""Returns base at position i in sequence identified by seq_id.
Args:
seq_id: ID of sequence present in given fasta file.
i: position of the base to check in fasta file.
Returns:
The base present in seuqence seq_id at position i.
"""
if seq_id not in self._sequences.keys():
raise FastaProviderException(f"Wrong sequence id: {seq_id}. ")
if i > len(self._sequences[seq_id]):
raise FastaProviderException(f"""Index {i} is too large
for sequence {seq_id}.""")
return graph.Base(self._sequences[seq_id][i])
def test_2_read_seqeunce_from_cache_instead_downloading(self):
fasta_provider = missings.FromNCBI(use_cache=True)
cache_dir_path = pathtools.get_child_path(Path.cwd(), ".fastacache")
if cache_dir_path.exists():
shutil.rmtree(cache_dir_path)
cache_dir_path.mkdir()
sequence_id = msa.SequenceID("seq1")
fake_sequence = "foo"
expected_base = graph.Base("o")
fake_fasta_path = pathtools.get_child_path(cache_dir_path,
f"{sequence_id}.fasta")
with open(fake_fasta_path, 'w') as fake_fasta_handler:
fake_fasta_handler.write(f">{sequence_id} cached\n{fake_sequence}")
actual_base = fasta_provider.get_base(sequence_id, 2)
self.assertEqual(expected_base, actual_base)
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_base(self, seq_id: msa.SequenceID, i: int) -> graph.Base:
"""Returns base at position i in sequence identified in NCBI
by seq_id or sth similar.
Args:
seq_id: sequence_id, used as is or some guessing is performed
if no results available.
i: position of the base to check in fasta file.
Returns:
The base present in seuqence sequence_id at position i.
"""
if seq_id not in self._sequences.keys():
sequence_is_cached = self._fasta_disk_cache.seq_is_cached(seq_id)
if self._use_cache and sequence_is_cached:
self._sequences[seq_id] = self._fasta_disk_cache.read(seq_id)
elif self._use_cache and not sequence_is_cached:
sequence = self._download_from_ncbi(seq_id)
self._sequences[seq_id] = sequence
self._fasta_disk_cache._save_to_cache(seq_id, sequence)
else:
self._sequences[seq_id] = self._download_from_ncbi(seq_id)
return graph.Base(self._sequences[seq_id][i])
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 __init__(self, missing_base: MissingBase):
self.missing_base: graph.Base = graph.Base(missing_base.value)
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_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_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_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 b(x):
return graph.Base(x)
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_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)
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 bid(x):
return graph.Base(x)
def get_base(self, sequence_id: msa.SequenceID, i: int) -> graph.Base:
try:
return graph.Base(self.sources[sequence_id][i])
except KeyError:
raise Exception("No record found with given id_!")
