# print("\t#", scaffold_name)
print("#", scaffold_name, file=target)
# print("\t", " ".join(
# strand + str(gene_name) for strand, gene_name in grimm_formatted_genomes[genome][scaffold_name]),
# "$")
print(" ".join(strand + str(gene_name) for strand, gene_name in
grimm_formatted_genomes[genome][scaffold_name]),
"$",
file=target)
# print()
bg_graphs = dict()
for file_name in os.listdir(target_directory):
file_name = os.path.join(target_directory, file_name)
with open(file_name, "r") as source:
bg = GRIMMReader.get_breakpoint_graph(source)
bg_graphs[file_name.split(".")[0]] = bg
bg = BreakpointGraph()
for br_gr in bg_graphs.values():
bg.update(br_gr, merge_edges=True)
target_multicolor = Multicolor("Anguilla_japonica")
print("Breakpoint graph stats:")
print(
"\t", "non-infinity nodes count:",
len(
list(node for node in bg.nodes()
if not BGVertex.is_infinity_vertex(node))))
normal_edges, infinity_edges = [], []
for edge in bg.edges():
################################################################################################################
#
# END OF experiment set up (data)
#
################################################################################################################
if __name__ == "__main__":
length = sys.argv[1]
identity = sys.argv[2]
for i in range(0, len(TARGET_ORGANISM_NAMES)):
GRIMM_FILES.append(PATH + length + "_" + identity + "/" + TARGET_ORGANISM_NAMES[i] + ".txt")
print("Reading data into breakpoint graph...")
graph = BreakpointGraph()
for file in GRIMM_FILES:
with open(file, "rt") as source:
graph.update(GRIMMReader.get_breakpoint_graph(source), merge_edges=True)
print("Getting a tree...")
bgtree = NewickReader.from_string(NEWICK_STRING_TREE)
print("Preparing organisms for assembling...")
target_organisms = [BGGenome(organism) for organism in TARGET_ORGANISM_NAMES]
exclude = [BGGenome(organism) for organism in COMPLETE_ORGANISM_NAMES]
print("Staring the assembly process...")
result = assemble_scaffolds(graph=graph, bgtree=bgtree, target_organisms=target_organisms, exclude=exclude,
verbose=True)
print("Finished assembling!")
print("Were identified",
sum([len(multicolor.colors) for v1, v2, weight, repeat_name, multicolor in result]),
"assembly points")
def get_html_report_experiment_entry(experiment):
try:
module_path, file_name = os.path.split(experiment.config_file_path)
if module_path not in sys.path:
sys.path.insert(0, module_path)
module_name = file_name[:file_name.rfind(".")]
if module_name in sys.modules:
del sys.modules[module_name]
module = importlib.import_module(module_name)
config = module.configuration
except Exception:
raise
config = recursive_dict_update(default_configuration.configuration, config)
result = []
result.append("<hr>")
result.append("<hr>")
result.append("<h1>{experiment_name}</h1>".format(experiment_name=config.get("experiment_name", "Experiment XXX")))
reference_chr_fragments_order_full_path = os.path.join(experiment.evaluation_dir_path, "chr_fragments_order.txt")
assembled_chains_full_path = os.path.join(config["gos-asm"]["output"]["dir"], config["gos-asm"]["output"]["chains_file"])
assembly_points_file_full_path = os.path.join(config["gos-asm"]["output"]["dir"], config["gos-asm"]["output"]["assembly_points_file"])
assembly_point_evaluation = []
with open(assembly_points_file_full_path, newline="") as csv_file:
reader = csv.reader(csv_file, delimiter="|")
headers = next(reader)
headers = [header.strip() for header in headers]
for row in reader:
entry = {key: value for key, value in zip(headers, row)}
ap = AssemblyPoint.from_assembly_points_file(separated_values=entry)
assembly_point_evaluation.append(AssemblyPointEvaluation(ap=ap))
genomes_ref_chr_fragments_orders = defaultdict(list)
current_genome = None
with open(reference_chr_fragments_order_full_path, "rt") as source:
for line in source:
if len(line.strip()) == 0 or line.strip().startswith("#"):
continue
elif line.strip().startswith(">"):
current_genome = BGGenome(line.strip()[1:])
genomes_ref_chr_fragments_orders[current_genome].append(line.strip())
elif current_genome is not None:
genomes_ref_chr_fragments_orders[current_genome].append(line.strip())
assembled_genome_fragments_orders = defaultdict(list)
current_genome = None
with open(assembled_chains_full_path, "rt") as source:
for line in source:
if len(line.strip()) == 0 or line.strip().startswith("#"):
continue
elif line.strip().startswith(">"):
current_genome = BGGenome(line.strip()[1:])
assembled_genome_fragments_orders[current_genome].append(line.strip())
elif current_genome is not None:
assembled_genome_fragments_orders[current_genome].append(line.strip())
bg = BreakpointGraph()
for file_path in [reference_chr_fragments_order_full_path, assembled_chains_full_path]:
with open(file_path, "rt") as source:
bg.update(GRIMMReader.get_breakpoint_graph(source, merge_edges=False))
scjs = {}
genomes_fragmentation_total = {}
target_genomes = [BGGenome(genome_name) for genome_name in config["gos-asm"]["input"]["target_organisms"]]
for genome in target_genomes:
assembled_chains = assembled_genome_fragments_orders[genome]
reference_chains = genomes_ref_chr_fragments_orders[genome]
source = assembled_chains + reference_chains
bg = GRIMMReader.get_breakpoint_graph(source, merge_edges=False)
scjs[genome] = single_cut_and_join_distance(bg)
reference_assembly_graph = {}
current_genome = None
with open(reference_chr_fragments_order_full_path, "rt") as source:
for line in source:
if len(line.strip()) == 0 or line.strip().startswith("#"):
continue
elif line.strip().startswith(">"):
current_genome = BGGenome(line.strip()[1:])
if current_genome not in reference_assembly_graph:
reference_assembly_graph[current_genome] = Graph()
genomes_fragmentation_total[current_genome] = 0
elif current_genome is not None:
fragments = line.strip().split()
*fragments, chr_type = fragments
genomes_fragmentation_total[current_genome] += len(fragments) - 1
for fragment1, fragment2 in zip(fragments[:-1], fragments[1:]):
v1, v2 = get_pair_of_fragments_vertices(fragment1, fragment2)
reference_assembly_graph[current_genome].add_edge(v1, v2, attr_dict={"type": "HC"})
if chr_type == "@":
v1, v2 = get_pair_of_fragments_vertices(fragments[-1], fragments[0])
reference_assembly_graph[current_genome].add_edge(v1, v2, attr_dict={"type": "HC"})
for cnt, fragment in enumerate(fragments):
closure_fragments = get_closure_fragments(fragments, cnt, chr_type)
for fr in closure_fragments:
v1, v2 = get_pair_of_fragments_vertices(fragment, fr)
if not reference_assembly_graph[current_genome].has_edge(v1, v2):
reference_assembly_graph[current_genome].add_edge(v1, v2, attr_dict={"type": "GOC"})
for ap_eval in assembly_point_evaluation:
fragment1 = ap_eval.ap.fragment1
fragment2 = ap_eval.ap.fragment2
fragment1_sign = ap_eval.ap.fragment1_sign
fragment2_sign = ap_eval.ap.fragment2_sign
fr1_suffix = "h" if fragment1_sign == "+" else "t"
fr2_suffix = "t" if fragment2_sign == "+" else "h"
v1, v2 = fragment1 + fr1_suffix, fragment2 + fr2_suffix
graph = reference_assembly_graph[ap_eval.ap.info.target_color.colors.pop()]
if graph.has_edge(v1, v2):
type_ = graph[v1][v2]["type"]
if type_ == "HC":
ap_eval.HC = True
ap_eval.GOC = True
elif type_ == "GOC":
ap_eval.GOC = True
exp_id = config.get("experiment_name", "Experiment XXX")
exp_id = "_".join(exp_id.split())
result.append("<h2>Experiment info</h2>")
result.append("<div class='well'>")
result.append("<p>" + config.get("experiment_info", "No info").replace("\n", "<br>") + "</p>")
result.append("</div>")
result.append("<h3>Evaluation</h3>")
result.append("<h4>Per target genome</h4>")
for genome in target_genomes:
g_suffix = exp_id + "_" + genome.name
genome_ap = [ap_eval for ap_eval in assembly_point_evaluation if ap_eval.ap.info.target_color.colors.pop().name == genome.name]
add_collapse_header(heading=genome.name, suffix=g_suffix, result=result)
result.append("<div class='container' style='width:100%;'>")
add_collapse_header(heading="Assembly points", suffix=g_suffix + "_assembly_point", result=result, parent_data="collapse_{suffix}".format(suffix=g_suffix))
add_ap_table(genome_ap, result)
add_collapse_footer(result=result)
result.append("</div>")
result.append("<p>SCJ distance between produced assembly and a reference one: <b>{scj_distance}</b></p>".format(scj_distance=scjs[genome]))
result.append("<ul>")
total_cnt = len(genome_ap)
result.append("<li><p>Total # of identified assembly points: <b>{ap_cnt}</b></p></li>".format(ap_cnt=total_cnt))
hc_cnt = len([ap for ap in genome_ap if ap.HC])
result.append("<li><p>Correct assembly points: <b>{HC_cnt}</b></p></li>".format(HC_cnt=hc_cnt))
goc_cnt = len([ap for ap in genome_ap if ap.GOC and not ap.HC])
result.append(
"<li><p>Correct from Global Gene Order (GOC) perspective assembly points: <b>{GOC_cnt}</b></p></li>".format(GOC_cnt=goc_cnt))
incorrect_cnt = len([ap for ap in genome_ap if not ap.HC and not ap.GOC])
result.append("<li><p>Incorrect assembly points: <b>{ic_cnt}</b></p></li>".format(ic_cnt=incorrect_cnt))
result.append("</ul>")
# result.append("</div>")
result.append("<h5>Relative portions of identified assembly points</h5>")
result.append("<div class='progress'>")
for value, style in zip([hc_cnt, goc_cnt, incorrect_cnt], ["success", "info", "danger"]):
value_prs = value * 100 / total_cnt
result.append(
'<div class="progress-bar progress-bar-{style}" role="progressbar" aria-valuenow="{value}" aria-valuemin="0" aria-valuemax="100" style="width: {value}%">'
''.format(style=style, value=value_prs))
result.append("<span>{value:.2f}%</span>".format(value=value_prs, style=style))
result.append("</div>")
result.append("</div>")
overall_cnt = genomes_fragmentation_total[genome]
result.append("<h5>Absolute portions of identified assembly points</h5>")
result.append("<div class='progress'>")
for value, style in zip([hc_cnt, goc_cnt, incorrect_cnt], ["success", "info", "danger"]):
value_prs = value * 100 / overall_cnt
result.append(
'<div class="progress-bar progress-bar-{style}" role="progressbar" aria-valuenow="{value}" aria-valuemin="0" aria-valuemax="100" style="width: {value}%">'
''.format(style=style, value=value_prs))
result.append("<span>{value:.2f}%</span>".format(value=value_prs, style=style))
result.append("</div>")
add_collapse_footer(result=result)
result.append("<h4>Overall results</h4>")
add_collapse_header(heading="Assembly_points", suffix=exp_id, result=result)
add_ap_table(assembly_point_evaluation, result)
add_collapse_footer(result=result)
# result.append("<div class='row'>")
result.append("<ul>")
total_cnt = len(assembly_point_evaluation)
result.append("<li><p>Total # of identified assembly points: <b>{ap_cnt}</b></p></li>".format(ap_cnt=total_cnt))
hc_cnt = len([ap for ap in assembly_point_evaluation if ap.HC])
result.append("<li><p>Correct assembly points: <b>{HC_cnt}</b></p></li>".format(HC_cnt=hc_cnt))
goc_cnt = len([ap for ap in assembly_point_evaluation if ap.GOC and not ap.HC])
result.append(
"<li><p>Correct from Global Gene Order (GOC) perspective assembly points: <b>{GOC_cnt}</b></p></li>".format(GOC_cnt=goc_cnt))
incorrect_cnt = len([ap for ap in assembly_point_evaluation if not ap.HC and not ap.GOC])
result.append("<li><p>Incorrect assembly points: <b>{ic_cnt}</b></p></li>".format(ic_cnt=incorrect_cnt))
result.append("</ul>")
result.append("<h5>Relative portions of identified assembly points</h5>")
result.append("<div class='progress'>")
for value, style in zip([hc_cnt, goc_cnt, incorrect_cnt], ["success", "info", "danger"]):
value_prs = value * 100 / total_cnt
result.append(
'<div class="progress-bar progress-bar-{style}" role="progressbar" aria-valuenow="{value}" aria-valuemin="0" aria-valuemax="100" style="width: {value}%">'
''.format(style=style, value=value_prs))
result.append("<span>{value:.2f}%</span>".format(value=value_prs, style=style))
result.append("</div>")
result.append("</div>")
overall_cnt = sum([value for key, value in genomes_fragmentation_total.items() if key in target_genomes])
result.append("<h5>Absolute portions of identified assembly points</h5>")
result.append("<div class='progress'>")
for value, style in zip([hc_cnt, goc_cnt, incorrect_cnt], ["success", "info", "danger"]):
value_prs = value * 100 / overall_cnt
result.append(
'<div class="progress-bar progress-bar-{style}" role="progressbar" aria-valuenow="{value}" aria-valuemin="0" aria-valuemax="100" style="width: {value}%">'
''.format(style=style, value=value_prs))
result.append("<span>{value:.2f}%</span>".format(value=value_prs, style=style))
result.append("</div>")
result.append("</div>")
return "\n".join(result)