1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
|
"""
This module will contain functions to be used in computation of the data used to
generate various kinds of heatmaps.
"""
from functools import reduce
from typing import Any, Dict, Sequence
from gn3.computations.slink import slink
from gn3.db.traits import retrieve_trait_data, retrieve_trait_info
from gn3.computations.correlations2 import compute_correlation
def export_trait_data(
trait_data: dict, strainlist: Sequence[str], dtype: str = "val",
var_exists: bool = False, n_exists: bool = False):
"""
Export data according to `strainlist`. Mostly used in calculating
correlations.
DESCRIPTION:
Migrated from
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/base/webqtlTrait.py#L166-L211
PARAMETERS
trait: (dict)
The dictionary of key-value pairs representing a trait
strainlist: (list)
A list of strain names
type: (str)
... verify what this is ...
var_exists: (bool)
A flag indicating existence of variance
n_exists: (bool)
A flag indicating existence of ndata
"""
def __export_all_types(tdata, strain):
sample_data = []
if tdata[strain]["value"]:
sample_data.append(tdata[strain]["value"])
if var_exists:
if tdata[strain]["variance"]:
sample_data.append(tdata[strain]["variance"])
else:
sample_data.append(None)
if n_exists:
if tdata[strain]["ndata"]:
sample_data.append(tdata[strain]["ndata"])
else:
sample_data.append(None)
else:
if var_exists and n_exists:
sample_data += [None, None, None]
elif var_exists or n_exists:
sample_data += [None, None]
else:
sample_data.append(None)
return tuple(sample_data)
def __exporter(accumulator, strain):
# pylint: disable=[R0911]
if strain in trait_data["data"]:
if dtype == "val":
return accumulator + (trait_data["data"][strain]["value"], )
if dtype == "var":
return accumulator + (trait_data["data"][strain]["variance"], )
if dtype == "N":
return accumulator + (trait_data["data"][strain]["ndata"], )
if dtype == "all":
return accumulator + __export_all_types(trait_data["data"], strain)
raise KeyError("Type `%s` is incorrect" % dtype)
if var_exists and n_exists:
return accumulator + (None, None, None)
if var_exists or n_exists:
return accumulator + (None, None)
return accumulator + (None,)
return reduce(__exporter, strainlist, tuple())
def trait_display_name(trait: Dict):
"""
Given a trait, return a name to use to display the trait on a heatmap.
DESCRIPTION
Migrated from
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/base/webqtlTrait.py#L141-L157
"""
if trait.get("db", None) and trait.get("trait_name", None):
if trait["db"]["dataset_type"] == "Temp":
desc = trait["description"]
if desc.find("PCA") >= 0:
return "%s::%s" % (
trait["db"]["displayname"],
desc[desc.rindex(':')+1:].strip())
return "%s::%s" % (
trait["db"]["displayname"],
desc[:desc.index('entered')].strip())
prefix = "%s::%s" % (
trait["db"]["dataset_name"], trait["trait_name"])
if trait["cellid"]:
return "%s::%s" % (prefix, trait["cellid"])
return prefix
return trait["description"]
def cluster_traits(traits_data_list: Sequence[Dict]):
"""
Clusters the trait values.
DESCRIPTION
Attempts to replicate the clustering of the traits, as done at
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/Heatmap.py#L138-L162
"""
def __compute_corr(tdata_i, tdata_j):
if tdata_i[0] == tdata_j[0]:
return 0.0
corr_vals = compute_correlation(tdata_i[1], tdata_j[1])
corr = corr_vals[0]
if (1 - corr) < 0:
return 0.0
return 1 - corr
def __cluster(tdata_i):
return tuple(
__compute_corr(tdata_i, tdata_j)
for tdata_j in enumerate(traits_data_list))
return tuple(__cluster(tdata_i) for tdata_i in enumerate(traits_data_list))
def heatmap_data(formd, search_result, conn: Any):
"""
heatmap function
DESCRIPTION
This function is an attempt to reproduce the initialisation at
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/Heatmap.py#L46-L64
and also the clustering and slink computations at
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/Heatmap.py#L138-L165
with the help of the `gn3.computations.heatmap.cluster_traits` function.
It does not try to actually draw the heatmap image.
PARAMETERS:
TODO: Elaborate on the parameters here...
"""
threshold = 0 # webqtlConfig.PUBLICTHRESH
cluster_checked = formd.formdata.getvalue("clusterCheck", "")
strainlist = [
strain for strain in formd.strainlist if strain not in formd.parlist]
genotype = formd.genotype
def __retrieve_traitlist_and_datalist(threshold, fullname):
trait = retrieve_trait_info(threshold, fullname, conn)
return (trait, retrieve_trait_data(trait, conn))
traits_details = [
__retrieve_traitlist_and_datalist(threshold, fullname)
for fullname in search_result]
traits_list = tuple(x[0] for x in traits_details)
traits_data_list = [x[1] for x in traits_details]
exported_traits_data_list = tuple(
export_trait_data(td, strainlist) for td in traits_data_list)
return {
"target_description_checked": formd.formdata.getvalue(
"targetDescriptionCheck", ""),
"cluster_checked": cluster_checked,
"slink_data": (
slink(cluster_traits(exported_traits_data_list))
if cluster_checked else False),
"sessionfile": formd.formdata.getvalue("session"),
"genotype": genotype,
"nLoci": sum(map(len, genotype)),
"strainlist": strainlist,
"ppolar": formd.ppolar,
"mpolar":formd.mpolar,
"traits_list": traits_list,
"traits_data_list": traits_data_list,
"exported_traits_data_list": exported_traits_data_list
}
def compute_heatmap_order(
slink_data, xoffset: int = 40, neworder: tuple = tuple()):
"""
Compute the data used for drawing the heatmap proper from `slink_data`.
This function tries to reproduce the creation and update of the `neworder`
variable in
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/Heatmap.py#L120
and in the `web.webqtl.heatmap.Heatmap.draw` function in GN1
"""
d_1 = (0, 0, 0) # returned from self.draw in lines 391 and 399. This is just a placeholder
def __order_maker(norder, slnk_dt):
if isinstance(slnk_dt[0], int) and isinstance(slnk_dt[1], int):
return norder + (
(xoffset+20, slnk_dt[0]), (xoffset + 40, slnk_dt[1]))
if isinstance(slnk_dt[0], int):
return norder + ((xoffset + 20, slnk_dt[0]), )
if isinstance(slnk_dt[1], int):
return norder + ((xoffset + d_1[0] + 20, slnk_dt[1]), )
return __order_maker(__order_maker(norder, slnk_dt[0]), slnk_dt[1])
return __order_maker(neworder, slink_data)
def retrieve_strains_and_values(orders, strainlist, traits_data_list):
"""
Get the strains and their corresponding values from `strainlist` and
`traits_data_list`.
This migrates the code in
https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/Heatmap.py#L215-221
"""
# This feels nasty! There's a lot of mutation of values here, that might
# indicate something untoward in the design of this function and its
# dependents ==> Review
strains = []
values = []
rets = []
for order in orders:
temp_val = traits_data_list[order[1]]
for i, strain in enumerate(strainlist):
if temp_val[i] is not None:
strains.append(strain)
values.append(temp_val[i])
rets.append([order, strains[:], values[:]])
strains = []
values = []
return rets
|
