Merge pull request #33 from msk-access/hotfix-plotting-bug

plotting tests, bug in genotype heatmap,

Dockerfile

@@ -7,7 +7,7 @@ FROM python:3.6-slim
 ARG BUILD_DATE
 ARG BUILD_VERSION
 ARG LICENSE="Apache-2.0"
-ARG BIOMETRICS_VERSION=0.2.2
+ARG BIOMETRICS_VERSION=0.2.4
 ARG VCS_REF
 ################## METADATA ########################
 LABEL org.opencontainers.image.vendor="MSKCC"

biometrics/VERSION

@@ -1 +1 @@
-0.2.3
+0.2.4

biometrics/genotype.py

@@ -53,14 +53,15 @@ def _plot_heatmap(self, data, outdir, name, title="Discordance calculations betw
                 customdata=data.to_numpy(),
                 hovertemplate='<b>Reference sample:</b> %{customdata[0]}' +
                               '<br><b>Query sample:</b> %{customdata[1]}' +
-                              '<br><b>Homozygous count in reference:</b> %{customdata[3]}' +
-                              '<br><b>Total match count:</b> %{customdata[4]}' +
-                              '<br><b>Homozygous match count:</b> %{customdata[5]}' +
-                              '<br><b>Heterozygous match count:</b> %{customdata[6]}' +
-                              '<br><b>Homozygous mismatch count:</b> %{customdata[7]}' +
-                              '<br><b>Heterozygous mismatch count:</b> %{customdata[8]}' +
-                              '<br><b>Discordance rate:</b> %{customdata[9]}' +
-                              '<br><b>Status:</b> %{customdata[12]}' +
+                              '<br><b>Count of common sites:</b> %{customdata[4]}' +
+                              '<br><b>Homozygous count in reference:</b> %{customdata[5]}' +
+                              '<br><b>Total match count:</b> %{customdata[6]}' +
+                              '<br><b>Homozygous match count:</b> %{customdata[7]}' +
+                              '<br><b>Heterozygous match count:</b> %{customdata[8]}' +
+                              '<br><b>Homozygous mismatch count:</b> %{customdata[9]}' +
+                              '<br><b>Heterozygous mismatch count:</b> %{customdata[10]}' +
+                              '<br><b>Discordance rate:</b> %{customdata[11]}' +
+                              '<br><b>Status:</b> %{customdata[14]}' +
                               '<extra></extra>',
                 zmin=self.zmin,
                 zmax=self.zmax,
@@ -96,18 +97,20 @@ def plot(self, data, outdir):
 
         # make plot for comparing input samples with each other
 
-        data_sub = data[~data['DatabaseComparison']].copy()
+        data_sub = data[~data['IsInputToDatabaseComparison']].copy()
+        del data_sub['IsInputToDatabaseComparison']
         data_sub['DiscordanceRate'] = data_sub['DiscordanceRate'].map(
             lambda x: round(x, 4))
 
         if data_sub.shape[0] > 1:
             self._plot_heatmap(
-                data_sub, outdir, name='genotype_comparison_input_only.html',
+                data_sub, outdir, name='genotype_comparison_input.html',
                 title="Discordance calculations between input samples")
 
         # make plot for comparing input samples with database samples
 
-        data_sub = data[data['DatabaseComparison']].copy()
+        data_sub = data[data['IsInputToDatabaseComparison']].copy()
+        del data_sub['IsInputToDatabaseComparison']
         data_sub['DiscordanceRate'] = data_sub['DiscordanceRate'].map(
             lambda x: round(x, 4))
 
@@ -238,7 +241,7 @@ def compare_samples(self, samples):
                 samples_input, samples_input, samples)
 
             for i in range(len(results)):
-                results[i]['DatabaseComparison'] = True
+                results[i]['IsInputToDatabaseComparison'] = False
             comparisons += results
 
         # for each input sample, compare with all the samples in the db
@@ -248,14 +251,15 @@ def compare_samples(self, samples):
                 samples_input, samples_db, samples)
 
             for i in range(len(results)):
-                results[i]['DatabaseComparison'] = True
+                results[i]['IsInputToDatabaseComparison'] = True
             comparisons += results
 
         comparisons = pd.DataFrame(comparisons)
 
         # compute discordance rate
 
         comparisons['DiscordanceRate'] = comparisons['HomozygousMismatch'] / (comparisons['HomozygousInRef'] + EPSILON)
+
         # data['DiscordanceRate'] = data['DiscordanceRate'].map(lambda x: round(x, 6))
         comparisons.loc[comparisons['HomozygousInRef'] < 10, 'DiscordanceRate'] = np.nan
 
@@ -277,9 +281,7 @@ def compare_samples(self, samples):
             ~comparisons['Matched'] & ~comparisons['ExpectedMatch'], 'Status'] = "Expected Mismatch"
 
         self.comparisons = comparisons[[
-            'ReferenceSample', 'ReferenceSampleGroup', 'QuerySample', 'QuerySampleGroup',
-            'CountOfCommonSites', 'HomozygousInRef', 'TotalMatch', 'HomozygousMatch',
-            'HeterozygousMatch', 'HomozygousMismatch',
+            'ReferenceSample', 'ReferenceSampleGroup', 'QuerySample', 'QuerySampleGroup', 'IsInputToDatabaseComparison', 'CountOfCommonSites', 'HomozygousInRef', 'TotalMatch', 'HomozygousMatch', 'HeterozygousMatch', 'HomozygousMismatch',
             'HeterozygousMismatch', 'DiscordanceRate', 'Matched',
             'ExpectedMatch', 'Status']]
 

tests/test_biometrics.py

@@ -13,6 +13,9 @@
 from biometrics.extract import Extract
 from biometrics.genotype import Genotyper
 from biometrics.sex_mismatch import SexMismatch
+from biometrics.minor_contamination import MinorContamination
+from biometrics.major_contamination import MajorContamination
+
 
 CUR_DIR = os.path.dirname(os.path.abspath(__file__))
 
@@ -224,7 +227,7 @@ def setUp(self, mock_args):
 
         self.args = get_args()
 
-    def test_sample_minor_contamination(self):
+    def test_minor_contamination(self):
         samples = get_samples(self.args, extraction_mode=False)
         samples = run_minor_contamination(self.args, samples)
 
@@ -236,14 +239,26 @@ def test_sample_minor_contamination(self):
             samples['test_sample1'].metrics['minor_contamination']['n_contributing_sites'], 1,
             msg='Count of contributing sites for minor contamination is wrong.')
 
-    def test_sample_major_contamination(self):
+    def test_plot_minor_contamination(self):
+        samples = get_samples(self.args, extraction_mode=False)
+        minor_contamination = MinorContamination(threshold=self.args.minor_threshold)
+        samples = minor_contamination.estimate(samples)
+        minor_contamination.plot(samples, self.args.outdir)
+
+    def test_major_contamination(self):
         samples = get_samples(self.args, extraction_mode=False)
         samples = run_major_contamination(self.args, samples)
 
         self.assertAlmostEqual(
             samples['test_sample1'].metrics['major_contamination']['val'], 0.2,
             places=1, msg='Major contamination is wrong.')
 
+    def test_plot_major_contamination(self):
+        samples = get_samples(self.args, extraction_mode=False)
+        major_contamination = MajorContamination(threshold=self.args.minor_threshold)
+        samples = major_contamination.estimate(samples)
+        major_contamination.plot(samples, self.args.outdir)
+
     def test_genotyper(self):
         samples = get_samples(self.args, extraction_mode=False)
 
@@ -258,6 +273,18 @@ def test_genotyper(self):
         self.assertEqual(len(data), 4, msg='There were not four comparisons done.')
         self.assertEqual(set(data['Status']), set(['Expected Match']), msg='All sample comparisons were expected to match.')
 
+    def test_genotyper_plot(self):
+        samples = get_samples(self.args, extraction_mode=False)
+
+        genotyper = Genotyper(
+            no_db_compare=self.args.no_db_compare,
+            discordance_threshold=self.args.discordance_threshold,
+            threads=self.args.threads,
+            zmin=self.args.zmin,
+            zmax=self.args.zmax)
+        data = genotyper.compare_samples(samples)
+        genotyper.plot(data, self.args.outdir)
+
     def test_sexmismatch(self):
         samples = get_samples(self.args, extraction_mode=False)