def calculatePrevalenceStatistic(data):
"""
Iterate over a state variable that has been created via the tabulateMarkerCounts
function. This will perform the necessary calculations to populate the state
variable with a prevalence statistic
Prevalence can be calculated by taking the total genotyped/CN and dividing it by
the sample size of the marker
"""
for dataElemList in generateCountList(data):
# If we are working with a 'genotype' or 'Genotyping failure' or
# 'No data' we want to skip prevalence calculations
if not validateGenotypes(list(dataElemList[2])):
continue
sampleSize = dataElemList[4]
markerGenotyped = dataElemList[5]
# If our genotyped count is 0 we want to set prevalence to 0
# to avoid division by zero
markerPrevalence = 0
if float(markerGenotyped) > 0:
markerPrevalence = float(markerGenotyped) / sampleSize
data[dataElemList[0]][dataElemList[1]][dataElemList[2]][dataElemList[3]]["prevalence"] = markerPrevalence
def incrementGenotypeCount(dict, metaKey, markerKey, genotype, groups, age):
"""
Increment the state dictionary with the three keys provided. If the key does
not already exist in the dictionary the default value is set to 1 otherwise
it is incremented by 1
If a group of ages is passed into this function we also want to categorize
all of our increments
"""
dict.setdefault(metaKey, OrderedDict()).setdefault(markerKey, OrderedDict()).setdefault(
genotype, OrderedDict()
).setdefault("All", OrderedDict()).setdefault("genotyped", 0)
genotypeAll = dict[metaKey][markerKey][genotype]["All"]["genotyped"]
genotypeAll += 1
# Initialize our sample size to 0 to avoid any errors
dict[metaKey][markerKey].setdefault("sample_size", OrderedDict()).setdefault("All", 0)
sampleAll = dict[metaKey][markerKey]["sample_size"]["All"]
if validateGenotypes(genotype):
sampleAll += 1
dict[metaKey][markerKey]["sample_size"]["All"] = sampleAll
dict[metaKey][markerKey][genotype]["All"]["genotyped"] = genotypeAll
# If our age key is not None we need to add this age group
if groups:
incrementCountsByAgeGroup(dict, metaKey, markerKey, genotype, groups, age)
def incrementCountsByAgeGroup(dict, metaKey, markerKey, genotype, groups, age):
"""
Initializes all age groups in our statistics dictionary and increments
only the age groups where a row of data containing that age was found
"""
groupKey = None
for group in groups:
# The groups list should contain a list of age groups in the following
# tuple format:
#
# [ (lower, upper, label), (lower, upper, label), .... ]
#
# We should always assume that our grouping will be lower <= age <= upper
# and our group key will be returned as "lower - upper".
#
# The two fringe cases we will have to look out for will be (0, upper)
# and (lower, 200) in these cases we are dealing with edge cases such as
# (0, 1) and (12, 200) which would be represented as age < 1 and
# age > 12
#
# The third element in the tuple, 'label', will represent the key assigned
# in the dictionary housing our statistics
(lower, upper, label) = group
dict[metaKey][markerKey]["sample_size"].setdefault(label, 0)
dict[metaKey][markerKey][genotype].setdefault(label, OrderedDict()).setdefault("genotyped", 0)
if age is not None:
if lower is None:
if float(age) < upper:
groupKey = label
if upper is None:
if float(age) > lower:
groupKey = label
if lower is not None and upper is not None:
if lower <= float(age) <= upper:
groupKey = label
if groupKey is not None:
# Once again, hacky but we do not want to increment the sample size for a given
# group if our genotype is 'Not genotyped' or 'Genotyping failure'
if validateGenotypes(genotype):
dict[metaKey][markerKey]["sample_size"][groupKey] += 1
dict[metaKey][markerKey][genotype][groupKey]["genotyped"] += 1