fix: Replace outdated df append

workflow/scripts/collect_lineage_calls.py

@@ -57,7 +57,7 @@ def collect_calls(sm_input, sm_output, states, lineage, number, length):
     ]
 
     # bring them together
-    call = pangolin_calls.append(call)
+    call = pd.concat([pangolin_calls, call])
 
     call.to_csv(sm_output, sep="\t", index=False)
 

workflow/scripts/evaluate-strain-call-error.py

@@ -62,7 +62,7 @@ def eval_error(paths, sm_output, max_reads, prefix, separator, percentage, load_
 
         df = df.merge(org_mix_df, how="outer").fillna(0)
 
-        results_df = results_df.append(df)
+        results_df = pd.concat([results_df, df])
 
     for sample in results_df["mix"].unique():
         sample_rmse = rmse(

workflow/scripts/extract-strains-from-gisaid-provision.py

@@ -21,7 +21,9 @@ def extract_strains_from_provision(
     chunks = pd.read_json(path_to_provision, lines=True, chunksize=9000)
     for i, chunk in enumerate(chunks):
         print(f"Parsing chunk {i}", file=sys.stderr)
-        provision = provision.append(select_oldest_strains(chunk), ignore_index=True)
+        provision = pd.concat(
+            [provision, select_oldest_strains(chunk)], ignore_index=True
+        )
     provision = select_oldest_strains(provision)
 
     # save strain genomes

workflow/scripts/generate-lineage-variant-table.py

@@ -75,21 +75,36 @@ def rename_enumeration(list_length):
             lineages = record.info["LINEAGES"]
             for signature in signatures:
                 # generate df with all signatures + VAF and Prob_not_present from calculation
-                variants_df = variants_df.append(
-                    {
-                        "Mutations": signature,
-                        "Frequency": vaf,
-                        "ReadDepth": dp,
-                        "Prob_not_present": prob_not_present,
-                    },
+                variants_df = pd.concat(
+                    [
+                        variants_df,
+                        pd.DataFrame(
+                            {
+                                "Frequency": vaf,
+                                "Mutations": signature,
+                                "Prob_not_present": prob_not_present,
+                                "ReadDepth": dp,
+                            },
+                            index=[0],
+                        ),
+                    ],
                     ignore_index=True,
                 )
-                # generate df with lineage matrix for all signatures
-                lineage_df = lineage_df.append(
-                    {
-                        "Mutations": signature,
-                        **{lineage.replace(".", " "): "x" for lineage in lineages},
-                    },
+
+                lineage_df = pd.concat(
+                    [
+                        lineage_df,
+                        pd.DataFrame(
+                            {
+                                "Mutations": [signature],
+                                **{
+                                    lineage.replace(".", " "): "x"
+                                    for lineage in lineages
+                                },
+                            },
+                            index=[0],
+                        ),
+                    ],
                     ignore_index=True,
                 )
 

workflow/scripts/generate-overview-table.py

@@ -65,7 +65,7 @@ def is_patient_report():
         columns=[eukaryota, bacteria, viruses, sars_cov2, unclassified]
     ).fillna(0)
     kraken_results["sample"] = sample
-    species_columns = species_columns.append(kraken_results, ignore_index=True)
+    species_columns = pd.concat([species_columns, kraken_results], ignore_index=True)
 
 data = data.join(species_columns.set_index("sample"))
 

workflow/scripts/plot-all-coverage.py

@@ -23,7 +23,7 @@ def plot_coverage(sm_input, sm_output, min_coverage):
 
         sample_df["Sample"] = sample_df["#CHROM"].apply(lambda x: str(x).split(".")[0])
 
-        coverage = coverage.append(sample_df, ignore_index=True)
+        coverage = pd.concat([coverage, sample_df], ignore_index=True)
 
     coverage["# Coverage"] = coverage.Coverage.apply(
         lambda x: f"< {min_coverage}"

workflow/scripts/plot-assembly-comparison.py

@@ -18,29 +18,43 @@ def register_lengths(sample, file_list, state, amplicon_state, data):
     for file, assembler in zip(file_list, snakemake.params.assembler):
         if state in ("initial", "scaffolded"):
             with pysam.FastxFile(file) as infile:
-                data = data.append(
-                    {
-                        "Sample": sample,
-                        "Assembler": assembler,
-                        "Amplicon": amplicon_state,
-                        "length (bp)": max(len(contig.sequence) for contig in infile),
-                        "State": state,
-                    },
+                data = pd.concat(
+                    [
+                        data,
+                        pd.DataFrame(
+                            {
+                                "Sample": sample,
+                                "Assembler": assembler,
+                                "Amplicon": amplicon_state,
+                                "length (bp)": max(
+                                    len(contig.sequence) for contig in infile
+                                ),
+                                "State": state,
+                            },
+                            index=[0],
+                        ),
+                    ],
                     ignore_index=True,
                 )
         else:
             quastDf = pd.read_csv(file, sep="\t")
-            data = data.append(
-                {
-                    "Sample": sample,
-                    "Assembler": assembler,
-                    "Amplicon": amplicon_state,
-                    "length (bp)": quastDf.loc[0, "N50"],
-                    "State": "N50",
-                    "Genome fraction (%)": quastDf.loc[0, "Genome fraction (%)"]
-                    if "Genome fraction (%)" in quastDf.columns
-                    else float("nan"),
-                },
+            data = pd.concat(
+                [
+                    data,
+                    pd.DataFrame(
+                        {
+                            "Sample": sample,
+                            "Assembler": assembler,
+                            "Amplicon": amplicon_state,
+                            "length (bp)": quastDf.loc[0, "N50"],
+                            "State": "N50",
+                            "Genome fraction (%)": quastDf.loc[0, "Genome fraction (%)"]
+                            if "Genome fraction (%)" in quastDf.columns
+                            else float("nan"),
+                        },
+                        index=[0],
+                    ),
+                ],
                 ignore_index=True,
             )
     return data

workflow/scripts/plot-dependency-of-pangolin-call.py

@@ -35,7 +35,7 @@ def plot_dependency_of_pangolin_call(sm_input, sm_output):
         pangolin_output["mixture_content"] = input.split(MIXTURE_PREFIX, 1)[-1].split(
             "."
         )[0]
-        all_sampes = all_sampes.append(pangolin_output, ignore_index=True)
+        all_sampes = pd.concat([all_sampes, pangolin_output], ignore_index=True)
 
     all_sampes["mixture_content"] = all_sampes["mixture_content"].str.replace("-", ".")
 

workflow/scripts/plot-pangolin-conflict.py

@@ -35,7 +35,7 @@ def plot_pangolin_conflict(sm_input, sm_output):
         pangolin_output = pd.read_csv(input)
         pangolin_output["true_lineage"] = true_lineage
         pangolin_output["true_lineage_percent"] = percent
-        all_sampes = all_sampes.append(pangolin_output, ignore_index=True)
+        all_sampes = pd.concat([all_sampes, pangolin_output], ignore_index=True)
 
     all_sampes["correct_lineage_assignment"] = (
         all_sampes["lineage"] == all_sampes["true_lineage"]

workflow/scripts/plot-primer-clipping.py

@@ -147,13 +147,13 @@ def plot_classes(counters):
     counts_before = count_intervals(file)
     counts_before["sample"] = sample
     counts_before["state"] = "before"
-    all_df = all_df.append(counts_before, ignore_index=True)
+    all_df = pd.concat([all_df, counts_before], ignore_index=True)
 
 for sample, file in iter_with_samples(snakemake.input.clipped):
     counts_after = count_intervals(file)
     counts_after["sample"] = sample
     counts_after["state"] = "after"
-    all_df = all_df.append(counts_after, ignore_index=True)
+    all_df = pd.concat([all_df, counts_after], ignore_index=True)
 
 bars, text = plot_classes(all_df)