def pmsignature_main(args):
import paplot.pmsignature as pmsignature
# config
[config,
conf_file] = tools.load_config(tools.win_to_unix(args.config_file))
input_list = tools.get_inputlist(tools.win_to_unix(args.input))
if len(input_list) == 0:
print("input no file.")
return
# dirs
output_html_dir = prep.create_dirs(tools.win_to_unix(args.output_dir),
args.project_name, config)
for input_file in input_list:
params_in = {
"dir": output_html_dir,
"data": input_file,
"project": args.project_name,
"title": args.title,
"ellipsis": args.ellipsis,
}
params_out = pmsignature.output_html(params_in, config)
if params_out is None:
continue
if params_out == {}:
prep.create_index(config,
tools.win_to_unix(args.output_dir),
"",
args.project_name,
args.title,
overview=args.overview,
sub_text="",
composite=True,
remarks=args.remarks)
continue
sig_num_sift = 0
if tools.config_getboolean(config, "result_format_pmsignature",
"background"):
sig_num_sift = 1
prep.create_index(config,
tools.win_to_unix(args.output_dir),
params_out["html"],
args.project_name,
args.title,
overview=args.overview,
sub_text="#sig %d" %
(params_out["sig_num"] + sig_num_sift),
composite=True,
remarks=args.remarks)
def create_html(dataset, params, config):
import os
import paplot.subcode.tools as tools
import paplot.prep as prep
html_div_template = "<div style='float: left;' id='div_pm{id}'></div>\n"
html_add_template = "msig_draw.add_div('div_pm{id}');\n"
div_text = ""
add_text = ""
for i in range(dataset["sig_num"]):
div_text += html_div_template.format(id=i)
add_text += html_add_template.format(id=i)
integral_text = ""
if dataset["intergral"] is True:
integral_text = html_integral_template
f_template = open(
os.path.dirname(os.path.abspath(__file__)) +
"/templates/graph_pmsignature.html")
html_template = f_template.read()
f_template.close()
sig_num_sift = 0
if tools.config_getboolean(config, "result_format_pmsignature",
"background"):
sig_num_sift = 1
f_html = open(params["dir"] + "/" + dataset["html"], "w")
f_html.write(
html_template.format(
project=params["project"],
title="%s(#sig %d)" %
(params["title"], dataset["sig_num"] + sig_num_sift),
data_js=dataset["js"],
version=prep.version_text(),
date=tools.now_string(),
divs=div_text,
add_divs=add_text,
integral=integral_text,
style="../style/%s" % os.path.basename(
tools.config_getpath(config, "style", "path", "default.js")),
))
f_html.close()
def load_subdata(ids, sec, config):
import os
import paplot.subcode.tools as tools
import paplot.convert as convert
import paplot.color as color
input_file = tools.config_getpath(config, sec, "path", default = "")
if os.path.exists(input_file) == False:
print ("[ERROR] file is not exist. %s" % input_file)
return None
sept = tools.config_getstr(config, sec, "sept").replace("\\t", "\t").replace("\\n", "\n").replace("\\r", "\r")
mode = tools.config_getstr(config, sec, "mode")
comment = tools.config_getstr(config, sec, "comment")
title = tools.config_getstr(config, sec, "title")
label = []
item = []
colors_n_di = {}
colors_h_di = {}
for name_set in tools.config_getstr(config, sec, "name_set").split(","):
name_set_split = convert.text_to_list(name_set, ":")
for i in range(len(name_set_split)):
text = name_set_split[i]
if i == 0:
item.append(text)
if len(name_set_split) == 1:
label.append(text)
elif i == 1:
label.append(text)
elif i == 2:
colors_n_di[name_set_split[0]] = color.name_to_value(text)
elif i == 3:
colors_h_di[name_set_split[0]] = color.name_to_value(text)
# fill in undefined items
colors_n_di = color.create_color_dict(item, colors_n_di, color.osaka_subway_colors)
colors_h_di2 = {}
for key in colors_n_di:
if key in colors_h_di: continue
colors_h_di2[key] = color.saturation_down(colors_n_di[key])
# dict to value
colors_n = []
for key in item:
colors_n.append(colors_n_di[key])
if mode == "range":
item.remove(item[0])
header = []
if tools.config_getboolean(config, sec, "header") == True:
pos_value = -1
pos_id = -1
else:
pos_value = tools.config_getint(config, sec, "col_value")-1
pos_id = tools.config_getint(config, sec, "col_ID")-1
header = ["",""]
# copy id_list for find check
unlookup = []
for iid in ids:
unlookup.append(iid)
# read
data_text = ""
values = []
for line in open(input_file):
line = line.strip()
if len(line.replace(sept, "")) == 0:
continue
if comment != "" and line.find(comment) == 0:
continue
if len(header) == 0:
header = convert.text_to_list(line,sept)
try:
colname = tools.config_getstr(config, sec, "col_value")
pos_value = header.index(colname)
colname = tools.config_getstr(config, sec, "col_ID")
pos_id = header.index(colname)
except Exception as e:
print(e.message)
return None
continue
cols = convert.text_to_list(line,sept)
if (cols[pos_id] in ids) == False: continue
else: unlookup.remove(cols[pos_id])
id_pos = ids.index(cols[pos_id])
if mode == "fix":
if cols[pos_value] in item:
data_text += subdata_data_template.format(id = id_pos, item = item.index(cols[pos_value]))
else:
print("[" + sec + "] name_set: data is undefined." + cols[pos_value] + "\n")
continue
elif mode == "range" or mode == "gradient":
try:
values.append(float(cols[pos_value]))
except Exception as e:
print(colname + ": data type is invalid.\n" + e.message)
continue
data_text += subdata_data_template.format(id = id_pos, item = cols[pos_value])
if len(unlookup) > 0:
print("[WARNING] can't find IDs subplot data.")
print(unlookup)
if mode == "gradient" and len(values) > 0:
item[0] = min(values)
item[1] = max(values)
return [data_text, item, colors_n, label, title]
def convert_tojs(params, config):
import os
import json
import paplot.subcode.tools as tools
import paplot.convert as convert
import paplot.color as color
# data read
try:
json_data = json.load(open(params["data"]))
except Exception as e:
print("failure open data %s, %s" % (params["data"], e.message))
return None
key_id_list = tools.config_getstr(config, "result_format_pmsignature",
"key_id")
key_ref = tools.config_getstr(config, "result_format_pmsignature",
"key_ref")
key_alt = tools.config_getstr(config, "result_format_pmsignature",
"key_alt")
key_strand = tools.config_getstr(config, "result_format_pmsignature",
"key_strand")
key_mutations = tools.config_getstr(config, "result_format_pmsignature",
"key_mutation")
key_mutation_count = tools.config_getstr(config,
"result_format_pmsignature",
"key_mutation_count")
sig_num = len(json_data[key_ref])
if sig_num == 0:
print("no data %s" % params["data"])
return {}
# signature names
signature_list = []
for s in range(sig_num):
signature_list.append("Signature %d" % (s + 1))
# each signature colors
sig_color_list = color.create_color_array(sig_num, color.r_set2)
# use background?
if tools.config_getboolean(config, "result_format_pmsignature",
"background"):
signature_list.append("Background ")
sig_color_list.append(color.r_set2_gray)
# Id list
id_txt = ""
if key_id_list in json_data:
id_txt = convert.list_to_text(json_data[key_id_list])
# mutations
mutations_txt = ""
if key_mutations in json_data:
for m in json_data[key_mutations]:
mutations_txt += "[%d,%d,%f]," % (m[0], m[1], m[2])
# signature
dataset_ref = ""
for sig in json_data[key_ref]:
tmp = ""
for sub in sig:
tmp += "[" + ",".join(map(str, sub)) + "],"
dataset_ref += ("[" + tmp + "],")
dataset_alt = ""
for sig in json_data[key_alt]:
tmp = ""
for sub in sig:
tmp += "[" + ",".join(map(str, sub)) + "],"
dataset_alt += ("[" + tmp + "],")
dataset_strand = ""
for sig in json_data[key_strand]:
dataset_strand += "[" + ",".join(map(str, sig)) + "],"
# tooltips
# for ref
keys_di = {
"a": "",
"c": "",
"g": "",
"t": "",
"ca": "",
"cg": "",
"ct": "",
"ta": "",
"tc": "",
"tg": "",
"plus": "",
"minus": "",
"id": "",
"sig": ""
}
tooltip_refs_txt = ""
for r in range(len(json_data[key_ref][0])):
tooltip_refs_txt += js_tooltip_ref_template.format(
index=r,
tooltip_format=convert.pyformat_to_jstooltip_text(
keys_di, config, "pmsignature", "", "tooltip_format_ref"))
mutation_count_txt = ""
if (key_mutation_count != "") and (key_mutation_count in json_data.keys()):
for v in json_data[key_mutation_count]:
mutation_count_txt += "%d," % v
# output
sig_num_sift = 0
if tools.config_getboolean(config, "result_format_pmsignature",
"background"):
sig_num_sift = 1
ellipsis = "%s%d" % (params["ellipsis"], (sig_num + sig_num_sift))
js_file = "data_%s.js" % ellipsis
html_file = "graph_%s.html" % ellipsis
f = open(params["dir"] + "/" + js_file, "w")
f.write(js_header + js_dataset.format(
Ids=id_txt,
color_A=tools.config_getstr(config, "pmsignature", "color_A",
"#06B838"),
color_C=tools.config_getstr(config, "pmsignature", "color_C",
"#609CFF"),
color_G=tools.config_getstr(config, "pmsignature", "color_G",
"#B69D02"),
color_T=tools.config_getstr(config, "pmsignature", "color_T",
"#F6766D"),
color_plus=tools.config_getstr(config, "pmsignature", "color_plus",
"#00BEC3"),
color_minus=tools.config_getstr(config, "pmsignature", "color_minus",
"#F263E2"),
signatures=convert.list_to_text(signature_list),
colors=convert.list_to_text(sig_color_list),
mutations=mutations_txt,
dataset_ref=dataset_ref,
dataset_alt=dataset_alt,
dataset_strand=dataset_strand,
tooltip_ref=tooltip_refs_txt,
tooltip_alt=convert.pyformat_to_jstooltip_text(
keys_di, config, "pmsignature", "", "tooltip_format_alt"),
tooltip_strand=convert.pyformat_to_jstooltip_text(
keys_di, config, "pmsignature", "", "tooltip_format_strand"),
mutation_title=convert.pyformat_to_jstooltip_text(
keys_di, config, "pmsignature", "",
"tooltip_format_mutation_title"),
mutation_partial=convert.pyformat_to_jstooltip_text(
keys_di, config, "pmsignature", "",
"tooltip_format_mutation_partial"),
mutation_count=mutation_count_txt,
))
f_template = open(
os.path.dirname(os.path.abspath(__file__)) +
"/templates/data_pmsignature.js")
js_function = f_template.read()
f_template.close()
f.write(js_function)
f.write(js_footer)
f.close()
integral = True
if key_id_list == "" or key_mutations == "" or key_mutation_count == "":
integral = False
return {
"sig_num": sig_num,
"js": js_file,
"html": html_file,
"intergral": integral,
}
def convert_tojs(params, config):
import json
import math
import itertools
import paplot.subcode.tools as tools
import paplot.convert as convert
import paplot.color as color
# data read
try:
jsonData = json.load(open(params["data"]))
except Exception as e:
print ("failure open data %s, %s" % (params["data"], e.message))
return None
key_Ids = tools.config_getstr(config, "result_format_signature", "key_id")
key_signature = tools.config_getstr(config, "result_format_signature", "key_signature")
key_mutations = tools.config_getstr(config, "result_format_signature", "key_mutation")
key_mutation_count = tools.config_getstr(config, "result_format_signature", "key_mutation_count")
sig_num = len(jsonData[key_signature])
if sig_num == 0:
print ("no data %s" % params["data"])
return {}
# signature names
signature_list = []
for s in range(sig_num):
signature_list.append("Signature %d" % (s+1))
# each signature colors
sig_color_list = color.create_color_array(sig_num, color.r_set2)
# use background?
if tools.config_getboolean(config, "result_format_signature", "background"):
signature_list.append("Background ")
sig_color_list.append(color.r_set2_gray)
# axis-y max
sig_y_max = tools.config_getint(config, "signature", "signature_y_max")
if (sig_y_max < 0):
for sig in jsonData[key_signature]:
for sub in sig:
m = max(sub)
if sig_y_max < m:
sig_y_max = m
# route list
sub_num = len(jsonData[key_signature][0][0])
log = math.log(sub_num, 4)
if log % 1 > 0:
print ("substitution's list length is invalid (%d, not number 4^N)" % sub_num)
return None
route_id = []
route_list = []
for p in itertools.product(("A","C","G","T"), repeat = int(log)):
route_id.append("".join(p))
route_list.append(p)
# substruction
sub_di = [
{"name":"C > A", "ref":"C", "color":tools.config_getstr(config, "signature", "alt_color_CtoA")},
{"name":"C > G", "ref":"C", "color":tools.config_getstr(config, "signature", "alt_color_CtoG")},
{"name":"C > T", "ref":"C", "color":tools.config_getstr(config, "signature", "alt_color_CtoT")},
{"name":"T > A", "ref":"T", "color":tools.config_getstr(config, "signature", "alt_color_TtoA")},
{"name":"T > C", "ref":"T", "color":tools.config_getstr(config, "signature", "alt_color_TtoC")},
{"name":"T > G", "ref":"T", "color":tools.config_getstr(config, "signature", "alt_color_TtoG")},
]
substruction = ""
for sub in sub_di:
route = []
for r in route_list:
route.append("p".join(r[0:int(log/2)]) + "p" + sub["ref"] + "p" + "p".join(r[int(log/2):]))
substruction += js_substruction_template.format(name = sub["name"], color = sub["color"], route = convert.list_to_text(route))
# Id list
id_txt = ""
if key_Ids in jsonData:
id_txt = convert.list_to_text(jsonData[key_Ids])
# mutations
mutations_txt = ""
if key_mutations in jsonData:
for m in jsonData[key_mutations]:
mutations_txt += "[%d,%d,%f]," % (m[0],m[1],m[2])
# signature
dataset_sig = ""
for sig in jsonData[key_signature]:
tmp = ""
for sub in sig:
tmp += "[" + ",".join(map(str, sub)) + "],"
dataset_sig += ("[" + tmp + "],")
mutation_count_txt = ""
if (key_mutation_count != "") and (key_mutation_count in jsonData.keys()):
for v in jsonData[key_mutation_count]:
mutation_count_txt += "%d," % v
# output
sig_num_sift = 0
if tools.config_getboolean(config, "result_format_signature", "background"):
sig_num_sift = 1
ellipsis = "%s%d" % (params["ellipsis"], (sig_num + sig_num_sift))
js_file = "data_%s.js" % ellipsis
html_file = "graph_%s.html" % ellipsis
keys_di = {"sig":"", "route":"", "id":""}
f = open(params["dir"] + "/" + js_file, "w")
f.write(js_header \
+ js_dataset.format(Ids = id_txt, \
signatures = convert.list_to_text(signature_list), \
colors = convert.list_to_text(sig_color_list), \
dataset_sig_max = sig_y_max, \
mutations = mutations_txt, \
dataset_sig = dataset_sig, \
route_id = convert.list_to_text(route_id), \
substruction = substruction, \
signature_title = convert.pyformat_to_jstooltip_text(keys_di, config, "signature", "", "tooltip_format_signature_title"), \
signature_partial = convert.pyformat_to_jstooltip_text(keys_di, config, "signature", "", "tooltip_format_signature_partial"), \
mutation_title = convert.pyformat_to_jstooltip_text(keys_di, config, "signature", "", "tooltip_format_mutation_title"), \
mutation_partial = convert.pyformat_to_jstooltip_text(keys_di, config, "signature", "", "tooltip_format_mutation_partial"), \
mutation_count = mutation_count_txt, \
)
+ js_function)
f.close()
integral = True
if key_Ids == "" or key_mutations == "" or key_mutation_count == "":
integral = False
return {"sig_num": sig_num,
"js": js_file,
"html": html_file,
"intergral": integral,
}
def load_subdata(ids, sec, config):
import os
import paplot.subcode.tools as tools
import paplot.convert as convert
import paplot.color as color
input_file = tools.config_getpath(config, sec, "path", default = "../../example/sample_summary.csv")
if os.path.exists(input_file) == False:
print ("[ERROR] file is not exist. %s" % input_file)
return None
sept = tools.config_getstr(config, sec, "sept")
mode = tools.config_getstr(config, sec, "mode")
comment = tools.config_getstr(config, sec, "comment")
title = tools.config_getstr(config, sec, "title")
label = []
item = []
colors_n_di = {}
colors_h_di = {}
for name_set in tools.config_getstr(config, sec, "name_set").split(","):
name_set_split = convert.text_to_list(name_set, ":")
for i in range(len(name_set_split)):
text = name_set_split[i]
if i == 0:
item.append(text)
if len(name_set_split) == 1:
label.append(text)
elif i == 1:
label.append(text)
elif i == 2:
colors_n_di[name_set_split[0]] = color.name_to_value(text)
elif i == 3:
colors_h_di[name_set_split[0]] = color.name_to_value(text)
# fill in undefined items
colors_n_di = color.create_color_dict(item, colors_n_di, color.osaka_subway_colors)
colors_h_di2 = {}
for key in colors_n_di:
if colors_h_di.has_key(key): continue
colors_h_di2[key] = color.Saturation_down(colors_n_di[key])
# dict to value
colors_n = []
for key in item:
colors_n.append(colors_n_di[key])
if mode == "range":
item.remove(item[0])
header = []
if tools.config_getboolean(config, sec, "header") == True:
pos_value = -1
pos_ID = -1
else:
pos_value = tools.config_getint(config, sec, "col_value")
pos_ID = tools.config_getint(config, sec, "col_ID")
header = ["",""]
# copy Ids for find check
unlookup = []
for iid in ids:
unlookup.append(iid)
# read
data_text = ""
values = []
for line in open(input_file):
line = line.strip()
if len(line.replace(sept, "")) == 0:
continue
if comment != "" and line.find(comment) == 0:
continue
if len(header) == 0:
header = convert.text_to_list(line,sept)
try:
colname = tools.config_getstr(config, sec, "col_value")
pos_value = header.index(colname)
colname = tools.config_getstr(config, sec, "col_ID")
pos_ID = header.index(colname)
except Exception as e:
print(e.message)
return None
continue
cols = convert.text_to_list(line,sept)
if (cols[pos_ID] in ids) == False: continue
else: unlookup.remove(cols[pos_ID])
id_pos = ids.index(cols[pos_ID])
if mode == "fix":
if cols[pos_value] in item:
data_text += subdata_data_template.format(id = id_pos, item = item.index(cols[pos_value]))
else:
print("[" + sec + "] name_set: data is undefined." + cols[pos_value] + "\n")
continue
elif mode == "range":
try:
values.append(float(cols[pos_value]))
except Exception as e:
print(colname + ": data type is invalid.\n" + e.message)
continue
data_text += subdata_data_template.format(id = id_pos, item = cols[pos_value])
elif mode == "gradient":
try:
values.append(float(cols[pos_value]))
except Exception as e:
print(colname + ": data type is invalid.\n" + e.message)
continue
data_text += subdata_data_template.format(id = id_pos, item = cols[pos_value])
if len(unlookup) > 0:
print("[WARNING] can't find IDs subplot data.")
print(unlookup)
if mode == "gradient" and len(values) > 0:
item[0] = min(values)
item[1] = max(values)
return [data_text, item, colors_n, label, title]
