def build_scaffolds(adjacencies, perm_container, debug_output,
correct_distances):
"""
Assembles scaffolds wrt to inferred adjacencies
"""
if debug_output:
logger.debug("Building scaffolds")
contigs, contig_index = _make_contigs(perm_container)
scaffolds = _extend_scaffolds(adjacencies, contigs, contig_index,
correct_distances)
num_contigs = sum([len(s.contigs) for s in scaffolds])
logger.debug("%d contigs were joined into %d scaffolds", num_contigs,
len(scaffolds))
if debugger.debugging and debug_output:
links_out = os.path.join(debugger.debug_dir, "scaffolder.links")
output_links(scaffolds, links_out)
contigs_out = os.path.join(debugger.debug_dir,
"scaffolder_contigs.txt")
output_permutations(perm_container.target_perms, contigs_out)
perms_out = os.path.join(debugger.debug_dir,
"scaffolder_scaffolds.txt")
output_scaffolds_premutations(scaffolds, perms_out)
return scaffolds
def merge_scaffolds(big_scaffolds, small_scaffolds, perm_container, rearrange):
"""
Merges scaffold sets from different iterations. If rearrangements are allowed,
tries to keep some small-scale rearrangements from the weaker scaffold set.
"""
logger.info("Merging two iterations")
#synchronizing scaffolds to the same permutations
big_updated = _update_scaffolds(big_scaffolds, perm_container)
small_updated = _update_scaffolds(small_scaffolds, perm_container)
if rearrange:
projector = RearrangementProjector(big_updated, small_updated, True)
new_adj = projector.project()
big_rearranged = build_scaffolds(new_adj, perm_container, False, False)
else:
big_rearranged = big_updated
merged_scf = _merge_scaffolds(big_rearranged, small_updated)
merged_scf = _merge_consecutive_contigs(merged_scf)
if debugger.debugging:
links_out = os.path.join(debugger.debug_dir, "merged.links")
output_links(merged_scf, links_out)
perms_out = os.path.join(debugger.debug_dir, "merged_scaffolds.txt")
output_scaffolds_premutations(merged_scf, perms_out)
return merged_scf
def build_scaffolds(adjacencies, perm_container, debug_output=True,
correct_distances=True, ancestral=False):
"""
Assembles scaffolds wrt to inferred adjacencies
"""
if debug_output:
logger.info("Building scaffolds")
if not ancestral:
contigs, contig_index = _make_contigs(perm_container)
else:
contigs, contig_index = _make_contigs(perm_container, ancestral=ancestral)
scaffolds = _extend_scaffolds(adjacencies, contigs, contig_index,
correct_distances)
num_contigs = sum(map(lambda s: len(s.contigs), scaffolds))
logger.debug("{0} contigs were joined into {1} scaffolds"
.format(num_contigs, len(scaffolds)))
if debugger.debugging and debug_output:
links_out = os.path.join(debugger.debug_dir, "scaffolder.links")
output_links(scaffolds, links_out)
contigs_out = os.path.join(debugger.debug_dir, "scaffolder_contigs.txt")
output_permutations(perm_container.target_perms, contigs_out)
perms_out = os.path.join(debugger.debug_dir, "scaffolder_scaffolds.txt")
output_scaffolds_premutations(scaffolds, perms_out)
return scaffolds
def merge_scaffolds(big_scaffolds, small_scaffolds, perm_container, rearrange, ancestral = False):
"""
Merges scaffold sets from different iterations. If rearrangements are allowed,
tries to keep some small-scale rearrangements from the weaker scaffold set.
"""
logger.info("Merging two iterations")
#synchronizing scaffolds to the same permutations
big_updated = _update_scaffolds(big_scaffolds, perm_container, ancestral=ancestral)
small_updated = _update_scaffolds(small_scaffolds, perm_container, ancestral=ancestral)
if rearrange:
projector = RearrangementProjector(big_updated, small_updated, True)
new_adj = projector.project()
big_rearranged = build_scaffolds(new_adj, perm_container, False, False, ancestral=ancestral)
else:
big_rearranged = big_updated
merged_scf = _merge_scaffolds(big_rearranged, small_updated)
merged_scf = _merge_consecutive_contigs(merged_scf)
if debugger.debugging:
links_out = os.path.join(debugger.debug_dir, "merged.links")
output_links(merged_scf, links_out)
perms_out = os.path.join(debugger.debug_dir, "merged_scaffolds.txt")
output_scaffolds_premutations(merged_scf, perms_out)
return merged_scf
def build_scaffolds(adjacencies,
perm_container,
debug_output=True,
correct_distances=True,
ancestral=False):
"""
Assembles scaffolds wrt to inferred adjacencies
"""
if debug_output:
logger.info("Building scaffolds")
if not ancestral:
contigs, contig_index = _make_contigs(perm_container)
else:
contigs, contig_index = _make_contigs(perm_container,
ancestral=ancestral)
scaffolds = _extend_scaffolds(adjacencies, contigs, contig_index,
correct_distances)
num_contigs = sum(map(lambda s: len(s.contigs), scaffolds))
logger.debug("{0} contigs were joined into {1} scaffolds".format(
num_contigs, len(scaffolds)))
if debugger.debugging and debug_output:
links_out = os.path.join(debugger.debug_dir, "scaffolder.links")
output_links(scaffolds, links_out)
contigs_out = os.path.join(debugger.debug_dir,
"scaffolder_contigs.txt")
output_permutations(perm_container.target_perms, contigs_out)
perms_out = os.path.join(debugger.debug_dir,
"scaffolder_scaffolds.txt")
output_scaffolds_premutations(scaffolds, perms_out)
return scaffolds
def do_job(recipe_file, out_dir, backend, assembly_refine,
overwrite, debug):
"""
Top-level logic of program
"""
out_log = os.path.join(out_dir, "ragout.log")
out_links = os.path.join(out_dir, "scaffolds.links")
out_scaffolds = os.path.join(out_dir, "scaffolds.fasta")
out_overlap = os.path.join(out_dir, "contigs_overlap.dot")
out_refined_links = os.path.join(out_dir, "scaffolds_refined.links")
out_refined_scaffolds = os.path.join(out_dir, "scaffolds_refined.fasta")
debug_root = os.path.join(out_dir, "debug")
if not os.path.isdir(out_dir):
os.mkdir(out_dir)
if debug:
debugger.set_debug_dir(debug_root)
debugger.clear_debug_dir()
enable_logging(out_log, debug)
logger.info("Cooking Ragout...")
if not check_extern_modules(backend):
return 1
try:
recipe = parse_ragout_recipe(recipe_file)
phylogeny = Phylogeny(recipe)
target_fasta_file = recipe["genomes"][recipe["target"]]["fasta"]
logger.info("Reading FASTA with contigs")
target_fasta_dict = read_fasta_dict(target_fasta_file)
except (RecipeException, FastaError, PhyloException) as e:
logger.error(e)
return 1
backends = SyntenyBackend.get_available_backends()
perm_files = backends[backend].make_permutations(recipe, out_dir, overwrite)
if not perm_files:
logger.error("There were problems with synteny backend, exiting.")
return 1
last_scaffolds = None
for block_size in recipe["blocks"]:
logger.info("Running Ragout with the block size {0}".format(block_size))
if debug:
debug_dir = os.path.join(debug_root, str(block_size))
debugger.set_debug_dir(debug_dir)
try:
perm_container = PermutationContainer(perm_files[block_size],
recipe)
except PermException as e:
logger.error(e)
return 1
graph = bg.BreakpointGraph()
graph.build_from(perm_container, recipe)
adjacencies = graph.find_adjacencies(phylogeny)
scaffolds = scfldr.get_scaffolds(adjacencies, perm_container)
if last_scaffolds:
last_scaffolds = merge.merge(last_scaffolds, scaffolds)
else:
last_scaffolds = scaffolds
debugger.set_debug_dir(debug_root)
out_gen.output_links(last_scaffolds, out_links)
out_gen.output_fasta(target_fasta_dict, last_scaffolds, out_scaffolds)
if assembly_refine:
if not ovlp.make_overlap_graph(target_fasta_file, out_overlap):
logger.error("Error in overlap graph reconstruction, exiting")
return 1
refined_scaffolds = asref.refine_scaffolds(out_overlap, last_scaffolds,
target_fasta_dict)
out_gen.output_links(refined_scaffolds, out_refined_links)
out_gen.output_fasta(target_fasta_dict, refined_scaffolds,
out_refined_scaffolds)
if debug:
shutil.copy(out_overlap, debugger.debug_dir)
out_colored_overlap = os.path.join(debugger.debug_dir,
"colored_overlap.dot")
asgraph.save_colored_insert_overlap_graph(out_overlap, last_scaffolds,
refined_scaffolds,
out_colored_overlap)
os.remove(out_overlap)
logger.info("Your Ragout is ready!")