Bases: ColocalisationMethodInterface
ECaviar-based colocalisation analysis.
It extends CAVIAR framework to explicitly estimate the posterior probability that the same variant is causal in 2 studies while accounting for the uncertainty of LD. eCAVIAR computes the colocalization posterior probability (CLPP) by utilizing the marginal posterior probabilities. This framework allows for multiple variants to be causal in a single locus.
Source code in src/gentropy/method/colocalisation.py
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 | class ECaviar(ColocalisationMethodInterface):
"""ECaviar-based colocalisation analysis.
It extends [CAVIAR](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5142122/#bib18) framework to explicitly estimate the posterior probability that the same variant is causal in 2 studies while accounting for the uncertainty of LD. eCAVIAR computes the colocalization posterior probability (**CLPP**) by utilizing the marginal posterior probabilities. This framework allows for **multiple variants to be causal** in a single locus.
"""
METHOD_NAME: str = "eCAVIAR"
METHOD_METRIC: str = "clpp"
@staticmethod
def _get_clpp(left_pp: Column, right_pp: Column) -> Column:
"""Calculate the colocalisation posterior probability (CLPP).
If the fact that the same variant is found causal for two studies are independent events,
CLPP is defined as the product of posterior porbabilities that a variant is causal in both studies.
Args:
left_pp (Column): left posterior probability
right_pp (Column): right posterior probability
Returns:
Column: CLPP
Examples:
>>> d = [{"left_pp": 0.5, "right_pp": 0.5}, {"left_pp": 0.25, "right_pp": 0.75}]
>>> df = spark.createDataFrame(d)
>>> df.withColumn("clpp", ECaviar._get_clpp(f.col("left_pp"), f.col("right_pp"))).show()
+-------+--------+------+
|left_pp|right_pp| clpp|
+-------+--------+------+
| 0.5| 0.5| 0.25|
| 0.25| 0.75|0.1875|
+-------+--------+------+
<BLANKLINE>
"""
return left_pp * right_pp
@classmethod
def colocalise(
cls: type[ECaviar],
overlapping_signals: StudyLocusOverlap,
**kwargs: Any,
) -> Colocalisation:
"""Calculate bayesian colocalisation based on overlapping signals.
Args:
overlapping_signals (StudyLocusOverlap): overlapping signals.
**kwargs (Any): Additional parameters passed to the colocalise method.
Returns:
Colocalisation: colocalisation results based on eCAVIAR.
"""
return Colocalisation(
_df=(
overlapping_signals.df.withColumns(
{
"clpp": ECaviar._get_clpp(
f.col("statistics.left_posteriorProbability"),
f.col("statistics.right_posteriorProbability"),
),
"tagVariantSource": get_tag_variant_source(f.col("statistics")),
}
)
.groupBy(
"leftStudyLocusId",
"rightStudyLocusId",
"rightStudyType",
"chromosome",
)
.agg(
# Count the number of tag variants that can be found in both loci:
f.size(
f.filter(
f.collect_list(f.col("tagVariantSource")),
lambda x: x == "both",
)
)
.cast(t.LongType())
.alias("numberColocalisingVariants"),
f.sum(f.col("clpp")).alias("clpp"),
)
.withColumn("colocalisationMethod", f.lit(cls.METHOD_NAME))
.join(
overlapping_signals.calculate_beta_ratio(),
on=["leftStudyLocusId", "rightStudyLocusId","chromosome"],
how="left"
)
),
_schema=Colocalisation.get_schema(),
)
|
colocalise(overlapping_signals: StudyLocusOverlap, **kwargs: Any) -> Colocalisation
classmethod
Calculate bayesian colocalisation based on overlapping signals.
Parameters:
| Name |
Type |
Description |
Default |
overlapping_signals
|
StudyLocusOverlap
|
|
required
|
**kwargs
|
Any
|
Additional parameters passed to the colocalise method.
|
{}
|
Returns:
| Name | Type |
Description |
Colocalisation |
Colocalisation
|
colocalisation results based on eCAVIAR.
|
Source code in src/gentropy/method/colocalisation.py
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 | @classmethod
def colocalise(
cls: type[ECaviar],
overlapping_signals: StudyLocusOverlap,
**kwargs: Any,
) -> Colocalisation:
"""Calculate bayesian colocalisation based on overlapping signals.
Args:
overlapping_signals (StudyLocusOverlap): overlapping signals.
**kwargs (Any): Additional parameters passed to the colocalise method.
Returns:
Colocalisation: colocalisation results based on eCAVIAR.
"""
return Colocalisation(
_df=(
overlapping_signals.df.withColumns(
{
"clpp": ECaviar._get_clpp(
f.col("statistics.left_posteriorProbability"),
f.col("statistics.right_posteriorProbability"),
),
"tagVariantSource": get_tag_variant_source(f.col("statistics")),
}
)
.groupBy(
"leftStudyLocusId",
"rightStudyLocusId",
"rightStudyType",
"chromosome",
)
.agg(
# Count the number of tag variants that can be found in both loci:
f.size(
f.filter(
f.collect_list(f.col("tagVariantSource")),
lambda x: x == "both",
)
)
.cast(t.LongType())
.alias("numberColocalisingVariants"),
f.sum(f.col("clpp")).alias("clpp"),
)
.withColumn("colocalisationMethod", f.lit(cls.METHOD_NAME))
.join(
overlapping_signals.calculate_beta_ratio(),
on=["leftStudyLocusId", "rightStudyLocusId","chromosome"],
how="left"
)
),
_schema=Colocalisation.get_schema(),
)
|