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(input_file, output_file, positions, config):
'''
Convert the input files to Json data and write them to the Javascript file
Also write functions and methods to process those data
Parameters
----------
input_file : str : The absolute path of formatted data file
output_file: str : The absolute path of JavaScript file
positions : dict: A nested dictionary with "must" and "option" as keys
config : configparser.RawConfigParser
Return
------
On success, return a dictionary: {"id_list": [...] "group_list": [...], "color": [...]}
id_list : The values for id column
group_list: The names of groups
color : The colors in groups
'''
import paplot.subcode.data_frame as data_frame
import paplot.subcode.merge as merge
import paplot.subcode.tools as tools
import paplot.convert as convert
import os
import math
# genome_size: a nested list
# [ [ A chromosome number in lowercase letters,
# The size of the 1st element,
# The color of the 1st element,
# The original name of the 1st element(that is not necessarily lowercase) or a user-defined name, ], ... ]
genome_size = load_genome_size(config)
if len(genome_size) == 0:
return None
# genome: dictionary-style string like this
# {"chr":"00", "size":249250621, "color":"#BBBBBB", "label":"1",},
# {"chr":"01", "size":243199373, "color":"#BBBBBB", "label":"2",},
# ...
# chr : Sequential number
# size : Size corresponding to the label
# color: Color corresponding to the label
# label: Name corresponding to chromosome
genome = ""
for i in range(len(genome_size)):
if len(genome) > 0:
genome += ",\n"
genome += genome_size_template.format(Chr=i, size=genome_size[i][1], color=genome_size[i][2], label=genome_size[i][3])
# Create a data frame that has title and data attributions
# title is a list like ['Break1', 'Break2', 'Chr1', 'Chr2', 'Sample']
# data is a nested list like [[16019088, 62784483, '14', '12', 'SAMPLE1'], ...]
try:
df = data_frame.load_file(
input_file, header=1,
sept=tools.config_getstr(config, "result_format_ca", "sept"),
comment=tools.config_getstr(config, "result_format_ca", "comment")
)
except Exception as e:
print("failure open data %s, %s" % (input_file, e.message))
return None
if len(df.data) == 0:
print("no data %s" % input_file)
return None
# Create groups, labels, and colors_n
# cols_di: a dictionary that merges must and option values
# : ex) {'chr1': 'Chr1', 'break1': 'Break1', 'chr2': 'Chr2', 'break2': 'Break2', 'id': 'Sample'}
cols_di = merge.position_to_dict(positions)
if "group" in cols_di:
for i in range(len(df.data)):
# A title may be stored in cols_di["group"]
group_pos = df.name_to_index(cols_di["group"]) # Get group(title) index
group = df.data[i][group_pos] # Get group(title) value for row i
# Modify group value
df.data[i][group_pos] = group.replace(" ", "_")
if group == "":
df.data[i][group_pos] = "_blank_"
# groups : list: group names
# labels : list: group names
# colors_n: list: color values for groups
[groups, colors_n] = convert.group_list(df.column(cols_di["group"]), "ca", "group", config)
labels = groups
else:
groups = ["outer", "inner"]
labels = ["Inter-chromosome", "Intra-chromosome"]
colors_n = ["#9E4A98", "#51BF69"] # purple, green
# Create group_text that is a dictionary-style string with name, label, color
conbined = []
for i in range(len(groups)):
conbined.append(group_template.format(name=groups[i], label=labels[i], color=colors_n[i]))
group_text = ",".join(conbined)
# id_list: Values for "id" column
# : Sorted without duplicates
id_list = []
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])] # iid: column value for "id" title
if iid != "":
id_list.append(iid)
id_list = list(set(id_list))
id_list.sort()
# option_keys: Store the option keys of the positions dictionary
option_keys = tools.dict_keys(cols_di) # option_keys: list: sorted keys of cols_di
option_keys.remove("id") # option key
option_keys.remove("chr1") # must key
option_keys.remove("break1") # must key
option_keys.remove("chr2") # must key
option_keys.remove("break2") # must key
if "group" in option_keys:
option_keys.remove("group") # option key
# node_size: Size to divide chromosomes
node_size_select = tools.config_getint(config, "ca", "selector_split_size", 5000000)
# Write header and dataset of JavaScript file
f = open(output_file, "w")
f.write(js_header + js_dataset.format(
node_size_detail=calc_node_size(genome_size, 500), # node size for detailed thumbnails
node_size_thumb=calc_node_size(genome_size, 250), # node size for rough thumbnails
node_size_select=node_size_select, # node size for bar graph
genome_size=genome, # A dictionary-style string containing keys of "chr", "size", "color", and "label"
IDs=convert.list_to_text(id_list), # A comma-separated string of id column values
group=group_text, # A dictionary-style string containing keys of "name", "label", and "color"
tooltip=convert.pyformat_to_jstooltip_text(cols_di, config, "ca", "result_format_ca", "tooltip_format"), # A dictionary-style string containing keys of "name", "label", and "color"
link_header=convert.list_to_text(option_keys),
))
# Write link of JavaScript file
f.write(js_links_1) # Write the leading part
data_links = []
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])] # iid: the value of "id" column
# Ignore empty string
if iid == "":
continue
chr1 = str(row[df.name_to_index(cols_di["chr1"])]) # chromosome1
pos1 = row[df.name_to_index(cols_di["break1"])] # break point1
chr2 = str(row[df.name_to_index(cols_di["chr2"])]) # chromosome2
pos2 = row[df.name_to_index(cols_di["break2"])] # break point2
# Check if chr1 and chr2 is in the genome list
# Check if pos1 and pos2 is in the chr1 length
# index1 and index2 are indexes of the genome_size for chr1 and chr2
[index1, rang] = insite_genome(genome_size, chr1, pos1)
if rang > 0:
print("breakpoint 1 is over range. chr%s: input=%d, range=%d" % (chr1, pos1, rang))
continue
if rang < 0:
#print("chr1 is undefined. %s" % (chr1))
continue
[index2, rang] = insite_genome(genome_size, chr2, pos2)
if rang > 0:
print("breakpoint 2 is over range. chr%s: input=%d, range=%d" % (chr2, pos2, rang))
continue
if rang < 0:
#print("chr2 is undefined. %s" % (chr2))
continue
# Whether chr1 and chr2 are the same chromosome
inner_flg = "false"
if (chr1 == chr2):
inner_flg = "true"
# Set group_id: -1, 0, 1, index values of groups
# : Sequential numbers identifying groups
group_id = -1 # Not belong to any groups
if "group" in cols_di:
# If the value of group column is in group list, then group_id is the index of the list
# Others, group_id is -1
group_id = convert.value_to_index(groups, row[df.name_to_index(cols_di["group"])], -1)
else:
if inner_flg == "false":
group_id = 0 # chr1 and chr2 are in the different group
else:
group_id = 1 # chr1 and chr2 are in the same group
# Add an element to data_links
data_links.append([iid, index1, pos1, index2, pos2, group_id])
# tooltip_items: Data for tooltip
tooltip_items = []
for k in range(len(option_keys)): # Loop in the column titles except group, id, and must keys (chr1, chr2, break1, and break2)
key = option_keys[k]
if cols_di[key] == "":
continue
tooltip_items.append(row[df.name_to_index(cols_di[key])])
# Write link
f.write(links_template.format(
ID=iid,
Chr1=index1, pos1=pos1, Chr2=index2, pos2=pos2,
inner_flg=inner_flg,
group_id=group_id,
tooltip="[" + convert.list_to_text(tooltip_items) + "],"))
f.write(js_links_2) # Write the ending part
# Write integral bar item
# link: [{bp1: iid, bp2: iid}, {...}, ...]
# : Separate elements by group_id
link = []
for g in range(len(groups)):
link.append({})
for dl in data_links:
# dl = [iid, index1, pos1, index2, pos2, group_id]
# iid : The value of id title
# index1/2: The index of genome_size
# pos1/2 : Bareak point
# group_id: Index of groups
# Chr: The index of genome_size
# Pos: A break position based on node
bp1 = "root.{Chr:0>2}.{Chr:0>2}_{Pos:0>3}".format(Chr=dl[1], Pos=int(math.floor(dl[2] / node_size_select)))
bp2 = "root.{Chr:0>2}.{Chr:0>2}_{Pos:0>3}".format(Chr=dl[3], Pos=int(math.floor(dl[4] / node_size_select)))
group_id = dl[5]
# For bp1
if bp1 not in link[group_id]:
link[group_id][bp1] = []
link[group_id][bp1].append(dl[0]) # Append iid
# For bp2
if bp1 != bp2:
if bp2 not in link[group_id]:
link[group_id][bp2] = []
link[group_id][bp2].append(dl[0]) # Append iid
select_value_text = ""
select_key_text = ""
select_item_text = ""
for i in range(len(groups)):
values = [] # [Number of id, ...]
keys = [] # [[genome_size index, Break position], ...]
items = [] # [[id_list index, ...], ...]
for bp in sorted(link[i].keys()):
# values element
# link[i][bp]: list that stores id column values at a break position of a chromosome in a group
# : Duplicate values are stored
values.append(len(link[i][bp]))
# keys element
parts = bp.split(".")[2].split("_") # parts: [Chr, Pos]
keys.append([int(parts[0]), int(parts[1])])
# items element
sort = sorted(list(set(link[i][bp]))) # Delete duplicates
temp = []
for t in sort:
temp.append(id_list.index(t)) # id_list that stores values of id column
items.append(temp)
select_value_text += "[%s]," % (",".join(map(str, values)).replace(" ", "")) # += [1,1,...],
select_key_text += "[%s]," % (",".join(map(str, keys)).replace(" ", "")) # += [[0,1],[0,25],...],
select_item_text += "[%s]," % (",".join(map(str, items)).replace(" ", "")) # += [[9],[8],...],
f.write(js_selection.format(
value=select_value_text,
key=select_key_text,
item=select_item_text
))
# Write rest of JavaScript file and footer
f_template = open(os.path.dirname(os.path.abspath(__file__)) + "/templates/data_ca.js") # ./templates/data_ca.js
js_function = f_template.read()
f_template.close()
f.write(js_function)
f.write(js_footer)
f.close()
return {"id_list": id_list, "group_list": groups, "color": colors_n}
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 convert_tojs(input_file, output_file, positions, config):
import paplot.subcode.data_frame as data_frame
import paplot.subcode.merge as merge
import paplot.subcode.tools as tools
import paplot.convert as convert
genome_size = load_genome_size(config)
if len(genome_size) == 0:
return None
genome = ""
for i in range(len(genome_size)):
if len(genome) > 0:
genome += ",\n"
genome += genome_size_template.format(Chr=i, size = genome_size[i][1], color = genome_size[i][2], label = genome_size[i][3])
cols_di = merge.position_to_dict(positions)
# data read
try:
df = data_frame.load_file(input_file, header = 1, \
sept = tools.config_getstr(config, "merge_format_sv", "sept"), \
comment = tools.config_getstr(config, "result_format_sv", "comment") \
)
except Exception as e:
print ("failure open data %s, %s" % (input_file, e.message))
return None
if len(df.data) == 0:
print ("no data %s" % input_file)
return None
# group list
if "group" in cols_di:
for f in range(len(df.data)):
group_pos = df.name_to_index(cols_di["group"])
group = df.data[f][group_pos]
df.data[f][group_pos] = group.replace(" ", "_")
if group == "":
df.data[f][group_pos] = "_blank_"
[groups, colors_n] = convert.group_list(df.column(cols_di["group"]), "sv", "group", config)
labels = groups
else:
groups = ["outer", "inner"]
labels = ["Inter Chromosome", "Intra Chromosome"]
colors_n = ["#9E4A98", "#51BF69"]
conbined = []
for i in range(len(groups)):
conbined.append(group_template.format(name = groups[i], label = labels[i], color = colors_n[i]))
group_text = ",".join(conbined)
# ID list
Ids = []
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])]
if iid != "": Ids.append(iid)
Ids = list(set(Ids))
Ids.sort()
option_keys = cols_di.keys()
option_keys.remove("id")
option_keys.remove("chr1")
option_keys.remove("break1")
option_keys.remove("chr2")
option_keys.remove("break2")
if "group" in option_keys:
option_keys.remove("group")
f = open(output_file, "w")
f.write(js_header \
+ js_dataset.format(node_size_detail = calc_node_size(genome_size, 500), \
node_size_thumb = calc_node_size(genome_size, 250), \
node_size_select = tools.config_getint(config, "sv", "selector_split_size", 5000000),\
genome_size = genome, \
IDs = convert.list_to_text(Ids), \
group = group_text, \
tooltip = convert.pyformat_to_jstooltip_text(cols_di, config, "sv", "result_format_sv", "tooltip_format"), \
link_header = convert.list_to_text(option_keys), \
))
# write links
f.write(js_links_1)
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])]
if iid == "": continue
chr1 = str(row[df.name_to_index(cols_di["chr1"])])
pos1 = row[df.name_to_index(cols_di["break1"])]
chr2 = str(row[df.name_to_index(cols_di["chr2"])])
pos2 = row[df.name_to_index(cols_di["break2"])]
[index1, rang] = insite_genome(genome_size, chr1, pos1)
if rang > 0:
print("breakpoint 1 is over range. chr%s: input=%d, range=%d" % (chr1, pos1, rang))
continue
if rang < 0:
#print("chr1 is undefined. %s" % (chr1))
continue
[index2, rang] = insite_genome(genome_size, chr2, pos2)
if rang > 0:
print("breakpoint 2 is over range. chr%s: input=%d, range=%d" % (chr2, pos2, rang))
continue
if rang < 0:
#print("chr2 is undefined. %s" % (chr2))
continue
inner_flg = "false"
if (chr1 == chr2):
inner_flg = "true"
tooltip_items = []
for k in range(len(option_keys)):
key = option_keys[k]
if cols_di[key] == "": continue
tooltip_items.append(row[df.name_to_index(cols_di[key])])
group_id = -1
if "group" in cols_di:
group_id = convert.value_to_index(groups, row[df.name_to_index(cols_di["group"])], -1)
else:
if inner_flg == "false":
group_id = 0
else:
group_id = 1
f.write(links_template.format(ID = iid, \
Chr1=index1, pos1=pos1, Chr2=index2, pos2=pos2, \
inner_flg = inner_flg, \
group_id = group_id , \
tooltip = "[" + convert.list_to_text(tooltip_items) + "],"))
f.write(js_links_2)
f.write(js_function)
f.close()
return {"id_list":Ids, "group_list":groups, "color":colors_n}
def convert_tojs(input_file, output_file, positions, config):
import paplot.subcode.data_frame as data_frame
import paplot.subcode.merge as merge
import paplot.subcode.tools as tools
import paplot.convert as convert
import math
genome_size = load_genome_size(config)
if len(genome_size) == 0:
return None
genome = ""
for i in range(len(genome_size)):
if len(genome) > 0:
genome += ",\n"
genome += genome_size_template.format(Chr=i,
size=genome_size[i][1],
color=genome_size[i][2],
label=genome_size[i][3])
cols_di = merge.position_to_dict(positions)
# data read
try:
df = data_frame.load_file(input_file, header = 1, \
sept = tools.config_getstr(config, "result_format_ca", "sept"), \
comment = tools.config_getstr(config, "result_format_ca", "comment") \
)
except Exception as e:
print("failure open data %s, %s" % (input_file, e.message))
return None
if len(df.data) == 0:
print("no data %s" % input_file)
return None
# group list
if "group" in cols_di:
for f in range(len(df.data)):
group_pos = df.name_to_index(cols_di["group"])
group = df.data[f][group_pos]
df.data[f][group_pos] = group.replace(" ", "_")
if group == "":
df.data[f][group_pos] = "_blank_"
[groups, colors_n] = convert.group_list(df.column(cols_di["group"]),
"ca", "group", config)
labels = groups
else:
groups = ["outer", "inner"]
labels = ["Inter-chromosome", "Intra-chromosome"]
colors_n = ["#9E4A98", "#51BF69"]
conbined = []
for i in range(len(groups)):
conbined.append(
group_template.format(name=groups[i],
label=labels[i],
color=colors_n[i]))
group_text = ",".join(conbined)
# ID list
Ids = []
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])]
if iid != "": Ids.append(iid)
Ids = list(set(Ids))
Ids.sort()
option_keys = tools.dict_keys(cols_di)
option_keys.remove("id")
option_keys.remove("chr1")
option_keys.remove("break1")
option_keys.remove("chr2")
option_keys.remove("break2")
if "group" in option_keys:
option_keys.remove("group")
# node_size
node_size_select = tools.config_getint(config, "ca", "selector_split_size",
5000000)
f = open(output_file, "w")
f.write(js_header \
+ js_dataset.format(node_size_detail = calc_node_size(genome_size, 500), \
node_size_thumb = calc_node_size(genome_size, 250), \
node_size_select = node_size_select,\
genome_size = genome, \
IDs = convert.list_to_text(Ids), \
group = group_text, \
tooltip = convert.pyformat_to_jstooltip_text(cols_di, config, "ca", "result_format_ca", "tooltip_format"), \
link_header = convert.list_to_text(option_keys), \
))
# write links
data_links = []
f.write(js_links_1)
for row in df.data:
iid = row[df.name_to_index(cols_di["id"])]
if iid == "": continue
chr1 = str(row[df.name_to_index(cols_di["chr1"])])
pos1 = row[df.name_to_index(cols_di["break1"])]
chr2 = str(row[df.name_to_index(cols_di["chr2"])])
pos2 = row[df.name_to_index(cols_di["break2"])]
[index1, rang] = insite_genome(genome_size, chr1, pos1)
if rang > 0:
print("breakpoint 1 is over range. chr%s: input=%d, range=%d" %
(chr1, pos1, rang))
continue
if rang < 0:
#print("chr1 is undefined. %s" % (chr1))
continue
[index2, rang] = insite_genome(genome_size, chr2, pos2)
if rang > 0:
print("breakpoint 2 is over range. chr%s: input=%d, range=%d" %
(chr2, pos2, rang))
continue
if rang < 0:
#print("chr2 is undefined. %s" % (chr2))
continue
inner_flg = "false"
if (chr1 == chr2):
inner_flg = "true"
tooltip_items = []
for k in range(len(option_keys)):
key = option_keys[k]
if cols_di[key] == "": continue
tooltip_items.append(row[df.name_to_index(cols_di[key])])
group_id = -1
if "group" in cols_di:
group_id = convert.value_to_index(
groups, row[df.name_to_index(cols_di["group"])], -1)
else:
if inner_flg == "false":
group_id = 0
else:
group_id = 1
data_links.append([iid, index1, pos1, index2, pos2, group_id])
f.write(links_template.format(ID = iid, \
Chr1=index1, pos1=pos1, Chr2=index2, pos2=pos2, \
inner_flg = inner_flg, \
group_id = group_id , \
tooltip = "[" + convert.list_to_text(tooltip_items) + "],"))
f.write(js_links_2)
# integral bar item
link = []
for g in range(len(groups)):
link.append({})
for l in data_links:
bp1 = "root.{Chr:0>2}.{Chr:0>2}_{Pos:0>3}".format(
Chr=l[1], Pos=int(math.floor(l[2] / node_size_select)))
bp2 = "root.{Chr:0>2}.{Chr:0>2}_{Pos:0>3}".format(
Chr=l[3], Pos=int(math.floor(l[4] / node_size_select)))
group = l[5]
#print group
# add bp1
if not bp1 in link[group]:
link[group][bp1] = []
link[group][bp1].append(l[0])
# add bp2
if bp1 != bp2:
if not bp2 in link[group]:
link[group][bp2] = []
link[group][bp2].append(l[0])
select_item_text = ""
select_value_text = ""
select_key_text = ""
for g in range(len(groups)):
items = []
values = []
keys = []
for i in link[g]:
values.append(len(link[g][i]))
# split key to chr and pos
parts = i.split(".")[2].split("_")
keys.append([int(parts[0]), int(parts[1])])
# delete duplication
sort = sorted(list(set(link[g][i])))
temp = []
for t in sort:
temp.append(Ids.index(t))
items.append(temp)
select_value_text += "[%s]," % (",".join(map(str, values)).replace(
" ", ""))
select_key_text += "[%s]," % (",".join(map(str, keys)).replace(
" ", ""))
select_item_text += "[%s]," % (",".join(map(str, items)).replace(
" ", ""))
f.write(
js_selection.format(value=select_value_text,
key=select_key_text,
item=select_item_text))
f.write(js_function)
f.close()
return {"id_list": Ids, "group_list": groups, "color": colors_n}
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]