def getTheGenes(theStartDate, templateList, dirr=os.getcwd()):
"""Walking the dir using Python 3.5. Variable theStartDate has to be
a datetime.date() data type."""
vv = ppma.lookForVARinList(templateList)
datOut = []
# dataOrdering = ['VAR', 'VARX', 'meanAllel', 'stdAllel', 'slope']
for dirName, subdirList, fileList in os.walk(dirr):
for file in fileList:
filepath = os.path.join(dirName, file)
if(filepath == os.path.join(dirName, 'InputParameters.json') and
ppma.loadTheDateFromParamFile(filepath) >= theStartDate):
paramzList = ppma.loadParamSettings(filepath)
# with open(filepath) as f:
# prms = json.load(f)
if ppma.compareParams(templateList, paramzList):
print("Data from:", dirName, end=" ")
popFiles = os.path.join(dirName, "HostGenomesFile.*.csv")
for fil in glob.glob(popFiles):
if not re.search('HostGenomesFile.0.csv', fil):
hostPopFile = fil
Gene_list = loadHostPopulation(hostPopFile)
geneStats = analiseGeneContent(Gene_list)
var = float(paramzList[vv['VAR']])
varx = float(paramzList[vv['VARX']])
geneStats['spp'] = varx
geneStats['patho_mut'] = var
datOut.append((geneStats))
print("- done!")
return datOut
def getTheData(theStartDate, templateList, dirr=os.getcwd()):
"""Walking the dir using Python 3.6. Variable theStartDate has to be
a datetime.date() data type."""
vv = ppma.lookForVARinList(templateList)
datOut = []
dataOrdering = ['VAR', 'VARX', 'slope', 'intercept']
for dirName, subdirList, fileList in os.walk(dirr):
for file in fileList:
filepath = os.path.join(dirName, file)
if (filepath == os.path.join(dirName, 'InputParameters.json') and
ppma.loadTheDateFromParamFile(filepath) >= theStartDate):
paramzList = ppma.loadParamSettings(filepath)
if ppma.compareParams(templateList, paramzList):
with open(filepath) as f:
prms = json.load(f)
path_spp = float(prms['number_of_pathogen_species'])
lg = prms['number_of_host_generations']
genomeFileName = "HostGenomesFile." + str(lg) + ".csv"
genomeFileName = os.path.join(dirName, genomeFileName)
# print(genomeFileName)
var = float(paramzList[vv['VAR']])
varx = float(paramzList[vv['VARX']])
try:
print(dirName, end=' : ')
pathos = loadPathoExposed(genomeFileName)
hosts = loadHostPopulation(genomeFileName)
if hosts is None or pathos is None:
print("Failed to read data")
continue
else:
print("Done")
except Exception:
print("ERROR in getTheData(): cant's load the host",
"population data")
continue
uniqNumb, pathoNumb = calculateTheNumbers(hosts, pathos)
uniqNumb = np.hstack((uniqNumb, 0))
pathoNumb = np.hstack((pathoNumb, 0))
# slope, intercept, r_val, p_val, std_err
data = linregress(uniqNumb, pathoNumb)
plotMHCvsPathoPresent(uniqNumb, pathoNumb, data[0],
data[1], dirName)
datOut.append((var, varx, data[0], data[1], data[2]**2,
data[3], data[4], path_spp, dirName))
datOut = np.array(datOut, dtype=outType)
return np.sort(datOut, order=dataOrdering)
def getTheData(theStartDate, templateList, dirr=os.getcwd()):
""" """
datOut = []
vv = ppma.lookForVARinList(templateList)
for dirName, subdirList, fileList in os.walk(dirr):
for file in fileList:
filepath = os.path.join(dirName, file)
if (filepath == os.path.join(dirName, 'InputParameters.json') and
ppma.loadTheDateFromParamFile(filepath) >= theStartDate):
paramzList = ppma.loadParamSettings(filepath)
if ppma.compareParams(templateList, paramzList):
var = float(paramzList[vv['VAR']])
varx = float(paramzList[vv['VARX']])
awkMeanINV(dirName)
meanINV = loadMeanInvdMhcNumb(dirName)
datOut.append((var, varx, meanINV))
print("Done dir:", dirName)
return datOut
def getTheData(theStartDate, templateList, dirr=os.getcwd(), genLast=0):
"""Walking the dir using Python 3.5. Variable theStartDate has to be
a datetime.date() data type. Each item in the `datOut` structure is the
result of computing one simulation."""
datOut = []
for dirName, subdirList, fileList in os.walk(dirr):
for file in fileList:
filepath = os.path.join(dirName, file)
if (filepath == os.path.join(dirName, 'InputParameters.json') and
ppma.loadTheDateFromParamFile(filepath) >= theStartDate):
paramzList = ppma.loadParamSettings(filepath)
if ppma.compareParams(templateList, paramzList):
print("Processing dir:", dirName, end=" ")
moPth = os.path.join(dirName, 'NumberOfMhcInMother.csv')
faPth = os.path.join(dirName, 'NumberOfMhcInFather.csv')
mPth = os.path.join(dirName, 'NumberOfMhcBeforeMating.csv')
mothr, fathr, bmate = loadTheParents(
genLast, moPth, faPth, mPth)
moth, fath, bmate = trimData(mothr, fathr, bmate, 2, 100)
mmMt = avrgMateMHCnumb(bmate)
rMom, rDad, rMmMt = reshapeMatherFather(moth, fath, mmMt)
ww, Fatrs, meanM = pickMotherSizeGroups(rMom, rDad, rMmMt)
bSize = np.zeros(len(Fatrs))
for i, itm in enumerate(Fatrs):
bSize[i] = len(itm)
deltas = []
for i, it in enumerate(ww):
deltas.append(np.nanmean(Fatrs[i] - meanM[i]))
justPlotDeviantFromMeanFather(ww, deltas, bSize, dirName)
try:
xx = np.transpose(
np.vstack((ww, np.array(deltas), bSize)))
except Exception:
print(" - failed to stack the data! Check if the",
"input file sizes (e.g. line numbers) are OK.")
continue
datOut.append(xx)
print(" - done.")
return datOut
def serchTheDirs(FILE, template, dirr=os.getcwd()):
"""Walk the directory tree in search of model runs and process each
simulation individually. Produces some meta-statistics regarding the
results geathered in Numpy structured array."""
# vv = ppma.lookForVAR(template)
vv = ppma.lookForVARinList(template)
datOut = []
dataOrdering = ['VAR', 'VARX', 'MRCA_time', 'maxMutNumb', 'numOfGenes']
for dirName, subdirList, fileList in os.walk(dirr):
for file in fileList:
filepath = os.path.join(dirName, file)
if filepath == os.path.join(dirName, FILE):
try:
paramList = ppma.loadParamSettings(os.path.join(dirName,
"InputParameters.json"))
except Exception:
print("Cannot load the parameters. in dir", dirName)
continue
if ppma.compareParams(template, paramList):
try:
DATA = processDataOneFile(filepath)
except Exception:
print("Cannot load the data. in dir", dirName)
continue
plotTheTimes(DATA[0], DATA[1], DATA[2], DATA[3], DATA[4],
dirName)
var = float(paramList[vv['VAR']])
varx = float(paramList[vv['VARX']])
datOut.append((var, varx, DATA[6], DATA[0].shape[1],
DATA[0].shape[0], dirName))
if len(datOut) > 0:
datOut = np.array(datOut, dtype=outType)
return np.sort(datOut, order=dataOrdering)
else:
print("ERROR in serchTheDirs(): output array is empty")
return None