reintroduce cohorts, aims, qc to metadata

Author: tristanpwdennis

malariagen_data/adir1.py

@@ -79,6 +79,7 @@ class Adir1(AnophelesDataResource):
     def __init__(
         self,
         url=None,
+        public_url=GCS_DEFAULT_PUBLIC_URL,
         bokeh_output_notebook=True,
         results_cache=None,
         log=sys.stdout,
@@ -94,6 +95,7 @@ def __init__(
     ):
         super().__init__(
             url=url,
+            public_url=public_url,
             config_path=CONFIG_PATH,
             cohorts_analysis=cohorts_analysis,
             aim_analysis=None,

malariagen_data/anoph/sample_metadata.py

@@ -621,28 +621,28 @@ def sample_metadata(
                 # Commented this out as it breaks some things - will fix
 
                 # Merge with the sequence QC metadata.
-                # df_sequence_qc = self.sequence_qc_metadata(
-                #    sample_sets=prepped_sample_sets
-                # )
+                df_sequence_qc = self.sequence_qc_metadata(
+                    sample_sets=prepped_sample_sets
+                )
 
                 # Note: merging can change column dtypes
-                # df_samples = df_samples.merge(
-                #    df_sequence_qc, on="sample_id", sort=False, how="left"
-                # )
+                df_samples = df_samples.merge(
+                    df_sequence_qc, on="sample_id", sort=False, how="left"
+                )
 
                 # If available, merge with the AIM metadata.
-                # if self._aim_analysis:
-                #    df_aim = self.aim_metadata(sample_sets=prepped_sample_sets)
-                #    df_samples = df_samples.merge(
-                #        df_aim, on="sample_id", sort=False, how="left"
-                #    )
+                if self._aim_analysis:
+                    df_aim = self.aim_metadata(sample_sets=prepped_sample_sets)
+                    df_samples = df_samples.merge(
+                        df_aim, on="sample_id", sort=False, how="left"
+                    )
 
                 # If available, merge with the cohorts metadata.
-                # if self._cohorts_analysis:
-                #    df_cohorts = self.cohorts_metadata(sample_sets=prepped_sample_sets)
-                #    df_samples = df_samples.merge(
-                #        df_cohorts, on="sample_id", sort=False, how="left"
-                #    )
+                if self._cohorts_analysis:
+                    df_cohorts = self.cohorts_metadata(sample_sets=prepped_sample_sets)
+                    df_samples = df_samples.merge(
+                        df_cohorts, on="sample_id", sort=False, how="left"
+                    )
 
             # Store sample metadata in the cache.
             self._cache_sample_metadata[cache_key] = df_samples

malariagen_data/anopheles.py

@@ -595,7 +595,17 @@ def roh_hmm(
         try:
             # Load cached numeric data, adding str / obj data again.
             results = self.results_cache_get(name=name, params=params)
-            df_roh = pd.DataFrame(results)
+
+            # Reconstruct dataframe
+            df_roh = pd.DataFrame(
+                {
+                    "roh_start": results["roh_start"],
+                    "roh_stop": results["roh_stop"],
+                    "roh_length": results["roh_length"],
+                    "roh_is_marginal": results["roh_is_marginal"],
+                }
+            )
+
             df_roh["sample_id"] = sample
             df_roh["contig"] = resolved_region.contig
 
@@ -630,6 +640,7 @@ def roh_hmm(
                 "roh_length",
                 "roh_is_marginal",
             ]
+
             self.results_cache_set(
                 name=name,
                 params=params,