Update UI (#2)

* Squashed 'pages/libs/mpxsonar/' content from commit 36581b5

pages/libs/mpxsonar/.env.template

@@ -1,3 +1,13 @@
+# DASH APP
+SERVER="0.0.0.0"
+PORT="80"
+SECRET_KEY="XX58602833XX"
+DEBUG=True
+
 # Database URL with generic format of connection.
 # https://USERNAME:PASSWORD@IP:PORT/DBNAME
 DB_URL="https://super_user:123456@localhost:3306/mpx"
+
+# Logging
+# DEBUG, INFO, WARNING, ERROR, CRITICAL
+LOG_LEVEL=DEBUG

pages/libs/mpxsonar/example-dash/app_controller.py

@@ -36,6 +36,7 @@ def match(
         with sonarDBManager(db, debug=debug) as dbm:
             if format == "vcf" and reference is None:
                 reference = dbm.get_default_reference_accession()
+
             cursor = dbm.match(
                 *profiles,
                 reserved_props=reserved_props_dict,
@@ -45,31 +46,29 @@ def match(
                 output_column=output_column,
                 showNX=showNX,
             )
-            # print(cursor)
             if format == "csv" or format == "tsv":
                 # cursor => list of dict
                 df = pd.DataFrame(cursor)
-                df.drop(["IMPORTED", "MODIFIED"], axis=1, inplace=True)
+                if "IMPORTED" in df.columns and "MODIFIED" in df.columns:
+                    df.drop(["IMPORTED", "MODIFIED"], axis=1, inplace=True)
                 # print(df)
-                output = df
-                tsv = True if format == "tsv" else False
-                sonarBasics.exportCSV(
-                    cursor, outfile=outfile, na="*** no match ***", tsv=tsv
-                )
+                if len(df) == 0:
+                    output = "*** no match ***"
+                else:
+                    output = df
+                # tsv = True if format == "tsv" else False
+                # sonarBasics.exportCSV(
+                #    cursor, outfile=outfile, na="*** no match ***", tsv=tsv
+                # )
             elif format == "count":
                 output = cursor.fetchone()["count"]
-                if outfile:
-                    with open(outfile, "w") as handle:
-                        handle.write(str(output))
-                else:
-                    print(output)
             elif format == "vcf":
+                # TODO: remove this. and change
                 sonarBasics.exportVCF(
                     cursor, reference=reference, outfile=outfile, na="*** no match ***"
                 )
             else:
                 sys.exit("error: '" + format + "' is not a valid output format")
-
         return output
 
     def list_prop(db=None):
@@ -120,15 +119,7 @@ def get_freq_mutation(args):
 def match_controller(args):  # noqa: C901
     props = {}
     reserved_props = {}
-    # for reserved keywords
-    reserved_key = ["sample"]
-    for pname in reserved_key:
-        if hasattr(args, pname):
-            if pname == "sample" and len(getattr(args, pname)) > 0:
-                # reserved_props[pname] = set([x.strip() for x in args.sample])
-                reserved_props = sonarBasics.set_key(
-                    reserved_props, pname, getattr(args, pname)
-                )
+
     with sonarDBManager(args.db, readonly=False, debug=args.debug) as dbm:
         for pname in dbm.properties:
             if hasattr(args, pname):
@@ -160,7 +151,7 @@ def match_controller(args):  # noqa: C901
                 )
             # reserved_props[pname] = getattr(args, pname)
     format = "count" if args.count else args.format
-
+    print(props)
     output = sonarBasicsChild.match(
         args.db,
         args.profile,

pages/libs/mpxsonar/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "mpxsonar"
-version = "0.3.7"
+version = "0.3.8"
 description = "A database-driven system for handling genomic sequences and screening genomic profiles."
 authors = ["Stephan Fuchs <FuchsS@rki.de>", "Kunaphas Kongkitimanon <KongkitimanonK@rki.de>", "Matthew Huska <HuskaM@rki.de>"]
 license = "GPL3"

pages/libs/mpxsonar/src/mpxsonar/dbm.py

@@ -1721,6 +1721,7 @@ def match(  # noqa: 901
                 # if 'sample' is presented we just use only samples
                 samples_condition = []
                 for pname, vals in reserved_props.items():
+                    print("sample" + str(vals))
                     if pname == "sample":
                         for x in vals:
                             samples_condition.append('"' + x + '"')
@@ -1741,6 +1742,7 @@ def match(  # noqa: 901
                         '"' + reference_accession + '"'
                     )
                 else:
+                    # should we use only sample or all alignment
                     sample_selection_sql = "SELECT id FROM sample"
                 if self.debug:
                     logging.info(f"sample_selection_sql: {sample_selection_sql}")
@@ -1838,7 +1840,7 @@ def match(  # noqa: 901
             _1_rows = self.cursor.fetchall()
 
             _2_final_sql = (
-                " SELECT name AS `sample.name`, nt_profile.reference_accession, nt_profile._profile AS NUC_PROFILE, aa_profile._profile AS AA_PROFILE \
+                " SELECT name AS `sample.name`, nt_profile.reference_accession AS REFERENCE_ACCESSION, nt_profile._profile AS NUC_PROFILE, aa_profile._profile AS AA_PROFILE \
                     FROM ( SELECT  `sample.id`, `reference.accession` AS reference_accession, group_concat("
                 + m
                 + "`variant.label`) AS _profile \
@@ -1906,7 +1908,14 @@ def match(  # noqa: 901
             # _1_rows = list(filter(None, _1_rows))
             # ------ alternative solution convert to df
             df_1 = pd.DataFrame(_1_rows)
+            df_1.sort_values(by=["sample.name"], inplace=True)
+            if self.debug:
+                logging.debug(df_1["sample.name"])
+            # sample.name REFERENCE_ACCESSION NUC_PROFILE AA_PROFILE
             df_2 = pd.DataFrame(_2_rows)
+            df_2.sort_values(by=["sample.name"], inplace=True)
+            if self.debug:
+                logging.debug(df_2["sample.name"])
             """
             merge_df = pd.merge(
                 df_1,
@@ -1917,18 +1926,26 @@ def match(  # noqa: 901
             )
 
             """
+            # [tmp solution.]- we remove unused column TODO: fix this in the future.
+            df_1.drop(
+                columns=["NUC_PROFILE", "AA_PROFILE", "NUC_N_PROFILE", "AA_X_PROFILE"],
+                inplace=True,
+            )
+            # NOTE: some samples might have only meta info, if we choose inner map
+            # the result will be intersection of two datafram.
             merge_df = pd.merge(
                 df_1,
                 df_2,
                 how="outer",
                 left_on=["sample.name"],
                 right_on=["sample.name"],
             )
-            # fix column for ref with noCDS
+
+            # fix column for ref with no CDS
             merge_df.loc[
-                merge_df["reference_accession"].isin(all_ref_nocds), "AA_PROFILE_y"
+                merge_df["REFERENCE_ACCESSION"].isin(all_ref_nocds), "AA_PROFILE"
             ] = "-"
-
+            merge_df.fillna("-", inplace=True)
             _1_rows = merge_df.to_dict("records")
             # filter column
             if output_column != "all":
@@ -1942,6 +1959,7 @@ def match(  # noqa: 901
             return _1_rows  # list(rows.values())
 
         elif format == "count":
+            logging.info("'--count' will return only unique sample.")
             sql = (
                 "SELECT COUNT(DISTINCT s2p.id) AS `count` FROM ("
                 + sample_selection_sql

pages/libs/mpxsonar/src/mpxsonar/sonar.py

@@ -696,7 +696,6 @@ def main(args):  # noqa: C901
                             reserved_props, "sample", line.strip()
                         )
         format = "count" if args.count else args.format
-
         sonarBasics.match(
             args.db,
             args.profile,