def test_change_samples_order(self):
test = pp.read_folder(path)
test.change_order(["Sample 4", "Sample 3", "Sample 2", "Sample 1"])
samples_test = ["Sample 4", "Sample 3", "Sample 2", "Sample 1"]
self.assertTrue(test.samples == samples_test)
comparison_test = [
'Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3',
'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3',
'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4'
]
self.assertTrue(test.comparisons == comparison_test)
self.assertEqual(len(test.genes_detected), 5)
self.assertEqual(len(test.Manipulate.significant(test, "gene")), 3)
self.assertEqual(len(test.isoforms_detected), 28)
self.assertEqual(len(test.Manipulate.significant(test, "isoform")), 5)
a = len(test.genes_detected.columns)
b = len(test.Manipulate.significant(test, "gene").columns)
c = len(test.isoforms_detected.columns)
d = len(test.Manipulate.significant(test, "isoform").columns)
self.assertTrue(a == b and b == c and c == d and d == 24)
with self.assertRaises(Exception):
test.change_order(["Sample 4", "Sample 3", "Sample 2"])
with self.assertRaises(Exception):
test.change_order(["Sample 4", "Sample 3", "Sample 2", "Wrong"])
test = pp.read_folder(path)
def drop(comp):
test2=pp.read_folder(path,drop_comparison=comp)
test3=pp.read_folder(path)
test3.dropComparison(comp)
df1=test2.genes_significant.all()
df2=test3.genes_significant.all()
self.assertTrue(df1.all()==df2.all())
def multidrop(comp):
test2=pp.read_folder(path)
test3=pp.read_folder(path)
test2.dropComparison(comp)
for c in comp:
test3.dropComparison(c)
df1=test2.genes_significant.all()
df2=test3.genes_significant.all()
self.assertTrue(df1.all()==df2.all())
def drop(comp):
test2 = pp.read_folder(path, drop_comparison=comp)
test3 = pp.read_folder(path)
test3.drop_comparison(comp)
df1 = test2.genes_detected.all()
df2 = test3.genes_detected.all()
self.assertTrue(df1.all() == df2.all())
df1 = test2.isoforms_detected.all()
df2 = test3.isoforms_detected.all()
self.assertTrue(df1.all() == df2.all())
def test_significant(self):
test = pp.read_folder(path)
self.assertEqual(len(test.Manipulate.significant(test, "gene")), 3)
self.assertEqual(len(test.Manipulate.significant(test, "isoform")), 5)
b = len(test.Manipulate.significant(test, "gene").columns)
d = len(test.Manipulate.significant(test, "isoform").columns)
self.assertTrue(b == d and d == 24)
def test_drop_comparison(self):
def drop(comp):
test2 = pp.read_folder(path, drop_comparison=comp)
test3 = pp.read_folder(path)
test3.drop_comparison(comp)
df1 = test2.genes_detected.all()
df2 = test3.genes_detected.all()
self.assertTrue(df1.all() == df2.all())
df1 = test2.isoforms_detected.all()
df2 = test3.isoforms_detected.all()
self.assertTrue(df1.all() == df2.all())
def multidrop(comp):
test2 = pp.read_folder(path)
test3 = pp.read_folder(path)
test2.drop_comparison(comp)
for c in comp:
test3.drop_comparison(c)
df1 = test2.genes_detected.all()
df2 = test3.genes_detected.all()
self.assertTrue(df1.all() == df2.all())
df1 = test2.isoforms_detected.all()
df2 = test3.isoforms_detected.all()
self.assertTrue(df1.all() == df2.all())
drop("Sample 1_vs_Sample 2")
drop("Sample 1_vs_Sample 3")
drop("Sample 1_vs_Sample 4")
drop("Sample 2_vs_Sample 3")
drop("Sample 2_vs_Sample 4")
drop("Sample 3_vs_Sample 4")
multidrop(["Sample 1_vs_Sample 2", "Sample 1_vs_Sample 3"])
multidrop(["Sample 1_vs_Sample 2", "Sample 3_vs_Sample 4"])
multidrop([
"Sample 1_vs_Sample 4", "Sample 2_vs_Sample 4",
"Sample 2_vs_Sample 3"
])
with self.assertRaises(Exception):
pp.read_folder(path, drop_comparison="Wrong")
test2 = pp.read_folder(path)
with self.assertRaises(Exception):
test2.drop_comparison("Wrong")
del test2
def test_selected_exist(self):
test2=pp.read_folder(path)
with self.assertRaises(Exception):
test2.selected_exist()
with self.assertRaises(Exception):
test2.selected_exist(remove="Wrong")
test2.get_gene()
self.assertTrue(test2.selected_exist())
del test2
def test_papillon_db(self):
test = pp.read_folder(path)
samples_test = ['Sample 1', 'Sample 2', 'Sample 3', 'Sample 4']
self.assertTrue(test.samples == samples_test)
comparison_test = [
'Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3',
'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3',
'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4'
]
self.assertTrue(test.comparisons == comparison_test)
self.assertEqual(len(test.genes_detected), 5)
self.assertEqual(len(test.isoforms_detected), 28)
a = len(test.genes_detected.columns)
c = len(test.isoforms_detected.columns)
self.assertTrue(a == c and c == 24)
print_test = pp.read_folder(path)
printable = "Samples: ['Sample 1', 'Sample 2', 'Sample 3', 'Sample 4']\nComparison: ['Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3', 'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3', 'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4']\nGenes Detected: 5\nGenes differential expressed: 3\nIsoform Detected: 28\nIsoform differential expressed: 5\n"
# print(print_test.__str__(),"\n",printable)
self.assertTrue(print_test.__str__() == printable)
del print_test
def test_read_folder(self):
test=pp.read_folder(path)
samples_test=['Sample 1', 'Sample 2', 'Sample 3', 'Sample 4']
self.assertTrue(test.samples==samples_test)
comparison_test=['Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3',
'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3',
'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4']
self.assertTrue(test.comparison==comparison_test)
self.assertEqual(len(test.genes_detect),5)
self.assertEqual(len(test.genes_significant),3)
self.assertEqual(len(test.isoforms_detect),28)
self.assertEqual(len(test.isoforms_significant),5)
a=len(test.genes_detect.columns)
b=len(test.genes_significant.columns)
c=len(test.isoforms_detect.columns)
d=len(test.isoforms_significant.columns)
self.assertTrue(a==b and b==c and c==d and d==18)
print_test=pp.read_folder(path)
printable="Samples: ['Sample 1', 'Sample 2', 'Sample 3', 'Sample 4']\nComparison: ['Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3', 'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3', 'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4']\nGenes Detected: 5\nGenes differential expressed: 3\nIsoform Detected: 28\nIsoform differential expressed: 5\nNone of the genes is selected"
print(print_test.__str__(),"\n",printable)
self.assertTrue(print_test.__str__()==printable)
del print_test
def test_different_read(self):
with self.assertRaises(FileNotFoundError):
pp.read_folder("Not working")
pp.read_folder(path)
pp.read_folder(path + "/galaxy")
pp.read_files([
path + "/gene_exp.diff", path + "/genes.fpkm_tracking",
path + "/isoform_exp.diff", path + "/isoforms.fpkm_tracking"
])
def test_list_search(self):
test = pp.read_folder(path)
sub = test.get_isoform()
sub_search = sub.search("sfd")
self.assertEqual(len(sub_search.df), 0)
sub_search = sub.search("00")
self.assertEqual(len(sub_search.df), 3)
self.assertEqual(len(sub.df.columns), len(sub_search.df.columns))
sub = test.get_gene()
sub_search = sub.search("sfd")
self.assertEqual(len(sub_search.df), 0)
sub_search = sub.search("il")
self.assertEqual(len(sub_search.df), 2)
self.assertEqual(len(sub.df.columns), len(sub_search.df.columns))
def heatmap_maker(z_score, type_sel):
test = pp.read_folder(path)
if type_sel == "gene":
sub = test.get_gene()
elif type_sel == "isoform":
sub = test.get_isoform()
df_heatmap = sub.onlyFPKM(return_as="gene name",
remove_FPKM_name=True)
df_heatmap = sub.plot._fusion_gene_id(df_heatmap,
type_sel,
change_index=True)
im1 = sns.clustermap(df_heatmap,
col_cluster=False,
method="complete",
cmap="seismic",
z_score=z_score)
im1.savefig(str(test.path + "test.png"))
def plot_maker(type_sel, z_score):
test = pp.read_folder(path)
if type_sel == "gene":
sub = test.get_gene()
elif type_sel == "isoform":
sub = test.get_isoform()
if z_score == True:
df_ = sub.onlyFPKM(return_as="df", remove_FPKM_name=True)
df_norm = sub.plot._z_score(df_)
df_norm["gene_short_name"] = sub.df["gene_short_name"]
df_ = df_norm.copy()
elif z_score == False:
df_ = sub.onlyFPKM(return_as="gene name",
remove_FPKM_name=True)
if type_sel == "gene":
hue = "gene_short_name"
df_ = sub.plot._fusion_gene_id(df_,
type_sel,
change_index=False)
elif type_sel == "isoform":
hue = "gene/ID"
df_ = sub.plot._fusion_gene_id(df_,
type_sel,
change_index=True)
df_ = df_.reset_index()
df = pd.melt(df_,
id_vars=hue,
var_name="Sample",
value_name="FPKM")
g = sns.factorplot(x="Sample",
y="FPKM",
hue=hue,
data=df,
ci=None,
legend=True,
size=10)
g.fig.suptitle(" Significant in AT LEAST one condition")
g.savefig(str(test.path + "test_plot.png"))
def test_add(self):
test = pp.read_folder(path)
# test isoform
sub1 = test.get_isoform("IL6")
sub2 = test.get_isoform("CD44")
sub = sub1 + sub2
self.assertEqual(len(sub.df), 3)
self.assertEqual(len(sub.df.columns), len(sub1.df.columns))
sub3 = test.get_isoform("CCL15")
sub = sum([sub1, sub2, sub3])
self.assertEqual(len(sub.df), 5)
self.assertEqual(len(sub.df.columns), len(sub1.df.columns))
sub4 = test.get_isoform("IL6")
sub = sub + sub4
self.assertEqual(len(sub.df), 5)
self.assertEqual(len(sub.df.columns), len(sub1.df.columns))
# test genes
sub_g1 = test.get_gene("IL6")
sub_g2 = test.get_gene("IL17RC")
sub = sub_g1 + sub_g2
self.assertEqual(len(sub.df), 2)
self.assertEqual(len(sub.df.columns), len(sub_g1.df.columns))
sub_g3 = test.get_gene("CCL15")
sub = sum([sub_g1, sub_g2, sub_g3])
self.assertEqual(len(sub.df), 3)
self.assertEqual(len(sub.df.columns), len(sub_g1.df.columns))
sub_g4 = test.get_gene("IL17RC")
sub = sub + sub_g4
self.assertEqual(len(sub.df), 3)
self.assertEqual(len(sub.df.columns), len(sub1.df.columns))
with self.assertRaises(Exception):
sub = sub1 + sub_g2
with self.assertRaises(Exception):
sub_g1 = test.get_gene("IL6")
sub_g2 = test.get_gene(comparison="Sample 3_vs_Sample 4")
sub = sub_g1 + sub_g2
def test_sub_select(self):
test = pp.read_folder(path)
sub = test.get_isoform()
a = sub.select("IL6")
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6"])
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6", "wrong"])
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6", "CCL15"])
self.assertEqual(len(a.df), 3)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
b = test.get_isoform(["IL6", "CCL15"])
a = sub.select(b)
self.assertEqual(len(a.df), 3)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
sub = test.get_gene()
a = sub.select("IL6")
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6"])
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6", "wrong"])
self.assertEqual(len(a.df), 1)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
a = sub.select(["IL6", "CCL15"])
self.assertEqual(len(a.df), 2)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
b = test.get_isoform(["IL6", "CCL15"])
a = sub.select(b)
self.assertEqual(len(a.df), 2)
self.assertEqual(len(a.df.columns), len(sub.df.columns))
def test_print(self):
test = pp.read_folder(path)
sub = test.get_gene()
printable2 = "Type of selection: gene\nNumber of gene selected: 3\nSamples: ['Sample 1', 'Sample 2', 'Sample 3', 'Sample 4']\nComparison selected: ['Sample 1_vs_Sample 2', 'Sample 1_vs_Sample 3', 'Sample 1_vs_Sample 4', 'Sample 2_vs_Sample 3', 'Sample 2_vs_Sample 4', 'Sample 3_vs_Sample 4']\n"
self.assertTrue(sub.__str__() == printable2)
def test_onlyFPKM(self):
test=pp.read_folder(path)
test.get_isoform()
df=test.onlyFPKM("df")
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),5)
self.assertEqual(len(df.columns),4)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df.index[-1],"NM_032965.4-2")
df=test.onlyFPKM("gene name")
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),5)
self.assertEqual(len(df.columns),5)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df["gene_short_name"][0],"IL6")
self.assertEqual(df.index[-1],"NM_032965.4-2")
self.assertEqual(df["gene_short_name"][-1],"CCL15")
df=test.onlyFPKM("array")
self.assertTrue(type(df)==numpy.ndarray)
self.assertEqual(len(df),5)
self.assertEqual(list(df[1]),[0.0, 3.0, 0.0, 0.0])
self.assertEqual(list(df[-1]),[0.0, 0.0, 0.0, 3.0])
#making extra_df
test.get_isoform()
extra_df=test.selected.iloc[:4,2:6].T.copy()
extra_df=pd.DataFrame(data=extra_df.values, index=test.selected.index[:4], columns=test.selected.columns[2:6])
extra_df['gene_short_name']=test.selected['gene_short_name'][:4]
#testing extra_df
df=test.onlyFPKM("df",extra_df=extra_df)
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),4)
self.assertEqual(len(df.columns),4)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df.index[-1],"NM_032965.4")
df=test.onlyFPKM("gene name",extra_df=extra_df)
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),4)
self.assertEqual(len(df.columns),5)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df["gene_short_name"][0],"IL6")
self.assertEqual(df.index[-1],"NM_032965.4")
self.assertEqual(df["gene_short_name"][-1],"CCL15")
df=test.onlyFPKM("array",extra_df=extra_df)
self.assertTrue(type(df)==numpy.ndarray)
self.assertEqual(len(df),4)
self.assertEqual(list(df[0]),[0.0, 0.0, 4.0, 0.0])
a=list(df[-1])
b=[0.016800, 0.0, 0.0, 0.0]
self.assertAlmostEqual(a[0],b[0], places=0)
#testing remove_FPKM_name
df=test.onlyFPKM("df",extra_df=extra_df, remove_FPKM_name=True)
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),4)
self.assertEqual(len(df.columns),4)
self.assertEqual(list(df.columns),test.samples)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df.index[-1],"NM_032965.4")
df=test.onlyFPKM("gene name",extra_df=extra_df, remove_FPKM_name=True)
self.assertTrue(type(df)==pd.DataFrame)
self.assertEqual(len(df),4)
self.assertEqual(len(df.columns),5)
self.assertEqual(list(df.columns[1:]),test.samples)
self.assertEqual(df.index[0],"NM_000600.3")
self.assertEqual(df["gene_short_name"][0],"IL6")
self.assertEqual(df.index[-1],"NM_032965.4")
self.assertEqual(df["gene_short_name"][-1],"CCL15")
# Final
test.get_isoform()
self.assertEqual(len(test.selected),5)
self.assertEqual(len(test.selected.columns),18)
import numpy
from scipy.stats import zscore
import unittest
import imagehash
from PIL import Image
sys.path.append(os.path.abspath(os.path.join('..')))
import papillon as pp
path_to_current_file = os.path.realpath(__file__)
current_directory = os.path.dirname(path_to_current_file)
os.chdir(current_directory)
path="Test_files"
test=pp.read_folder(path)
class papillon_Test(unittest.TestCase):
def test_different_read(self):
with self.assertRaises(FileNotFoundError):
pp.read_folder("Not working")
pp.read_folder(path)
pp.read_folder(path+"/galaxy")
pp.read_files([path+"/gene_exp.diff",path+"/genes.fpkm_tracking",path+"/isoform_exp.diff",path+"/isoforms.fpkm_tracking"])
# with self.assertWarns(DeprecationWarning):
# pp.read_db(path)
def test_functions_FPKM(self):
self.assertEqual(pp._FPKM("ciao"),"ciao_FPKM")
self.assertEqual(pp._FPKM("ciao_FPKM"),"ciao")
