From 7250ced981f2d19ab6001382075565775cd97466 Mon Sep 17 00:00:00 2001
From: GeorgWa <wallmann@biochem.mpg.de>
Date: Mon, 10 Jun 2024 18:03:58 -0400
Subject: [PATCH 1/5] use parquet as intermediate output

---
 alphadia/outputaccumulator.py |  4 ++--
 alphadia/outputtransform.py   | 16 ++++------------
 alphadia/planning.py          |  8 ++++----
 3 files changed, 10 insertions(+), 18 deletions(-)

diff --git a/alphadia/outputaccumulator.py b/alphadia/outputaccumulator.py
index 5328acab..f9465711 100644
--- a/alphadia/outputaccumulator.py
+++ b/alphadia/outputaccumulator.py
@@ -122,8 +122,8 @@ def parse_output_folder(
 
 
         """
-        psm_df = pd.read_csv(os.path.join(folder, "psm.tsv"), sep="\t")
-        frag_df = pd.read_csv(os.path.join(folder, "frag.tsv"), sep="\t")
+        psm_df = pd.read_parquet(os.path.join(folder, "psm.parquet"))
+        frag_df = pd.read_parquet(os.path.join(folder, "frag.parquet"))
 
         assert set(
             selected_precursor_columns
diff --git a/alphadia/outputtransform.py b/alphadia/outputtransform.py
index b62dc1cf..9276cff7 100644
--- a/alphadia/outputtransform.py
+++ b/alphadia/outputtransform.py
@@ -54,22 +54,14 @@ def get_frag_df_generator(folder_list: List[str]):
 
     for folder in folder_list:
         raw_name = os.path.basename(folder)
-        frag_path = os.path.join(folder, "frag.tsv")
+        frag_path = os.path.join(folder, "frag.parquet")
 
         if not os.path.exists(frag_path):
             logger.warning(f"no frag file found for {raw_name}")
         else:
             try:
                 logger.info(f"reading frag file for {raw_name}")
-                run_df = pd.read_csv(
-                    frag_path,
-                    sep="\t",
-                    dtype={
-                        "precursor_idx": np.uint32,
-                        "number": np.uint8,
-                        "type": np.uint8,
-                    },
-                )
+                run_df = pd.read_parquet(frag_path)
             except Exception as e:
                 logger.warning(f"Error reading frag file for {raw_name}")
                 logger.warning(e)
@@ -497,7 +489,7 @@ def build_precursor_table(
 
         for folder in folder_list:
             raw_name = os.path.basename(folder)
-            psm_path = os.path.join(folder, f"{self.PSM_INPUT}.tsv")
+            psm_path = os.path.join(folder, f"{self.PSM_INPUT}.parquet")
 
             logger.info(f"Building output for {raw_name}")
 
@@ -506,7 +498,7 @@ def build_precursor_table(
                 run_df = pd.DataFrame()
             else:
                 try:
-                    run_df = pd.read_csv(psm_path, sep="\t")
+                    run_df = pd.read_parquet(psm_path)
                 except Exception as e:
                     logger.warning(f"Error reading psm file for {raw_name}")
                     logger.warning(e)
diff --git a/alphadia/planning.py b/alphadia/planning.py
index 784028ff..f9fe1292 100644
--- a/alphadia/planning.py
+++ b/alphadia/planning.py
@@ -283,8 +283,8 @@ def run(
                 workflow_folder_list.append(workflow.path)
 
                 # check if the raw file is already processed
-                psm_location = os.path.join(workflow.path, "psm.tsv")
-                frag_location = os.path.join(workflow.path, "frag.tsv")
+                psm_location = os.path.join(workflow.path, "psm.parquet")
+                frag_location = os.path.join(workflow.path, "frag.parquet")
 
                 if self.config["general"]["reuse_quant"]:
                     if os.path.exists(psm_location) and os.path.exists(frag_location):
@@ -306,8 +306,8 @@ def run(
                     psm_df, frag_df = workflow.requantify_fragments(psm_df)
 
                 psm_df["run"] = raw_name
-                psm_df.to_csv(psm_location, sep="\t", index=False)
-                frag_df.to_csv(frag_location, sep="\t", index=False)
+                psm_df.to_parquet(psm_location, index=False)
+                frag_df.to_parquet(frag_location, index=False)
 
                 workflow.reporter.log_string(f"Finished workflow for {raw_name}")
                 workflow.reporter.context.__exit__(None, None, None)

From 64c9c94dcae5e47f2add19eea6e58fb001920b22 Mon Sep 17 00:00:00 2001
From: GeorgWa <wallmann@biochem.mpg.de>
Date: Mon, 10 Jun 2024 19:14:53 -0400
Subject: [PATCH 2/5] universal read write

---
 alphadia/consensus/utils.py        | 84 ++++++++++++++++++++++++++++++
 alphadia/constants/default.yaml    |  2 +
 alphadia/outputtransform.py        | 16 ++----
 alphadia/peakgroup/search.py       |  2 +-
 tests/unit_tests/conftest.py       | 12 +++++
 tests/unit_tests/test_consensus.py | 25 +++++++++
 6 files changed, 127 insertions(+), 14 deletions(-)
 create mode 100644 alphadia/consensus/utils.py
 create mode 100644 tests/unit_tests/test_consensus.py

diff --git a/alphadia/consensus/utils.py b/alphadia/consensus/utils.py
new file mode 100644
index 00000000..71054635
--- /dev/null
+++ b/alphadia/consensus/utils.py
@@ -0,0 +1,84 @@
+import logging
+import os
+import pandas as pd
+
+logger = logging.getLogger()
+supported_formats = ["parquet", "tsv"]
+
+
+def read_df(path_no_format, file_format="parquet"):
+    """Read dataframe from disk with choosen file format
+
+    Parameters
+    ----------
+
+    path_no_format: str
+        File to read from disk without file format
+
+    file_format: str, default = 'parquet'
+        File format for loading the file. Available options: ['parquet', 'tsv']
+
+    Returns
+    -------
+
+    pd.DataFrame
+        loaded dataframe from disk
+
+    """
+
+    if file_format not in supported_formats:
+        raise ValueError(
+            f"Provided unknown file format: {file_format}, supported_formats: {supported_formats}"
+        )
+
+    file_path = f"{path_no_format}.{file_format}"
+
+    if not os.path.exists(file_path):
+        raise FileNotFoundError(f"Can't load file as file was not found: {file_path}")
+
+    logger.info(f"Reading {file_path} from disk")
+
+    if file_format == "parquet":
+        return pd.read_parquet(file_path)
+
+    elif file_format == "tsv":
+        return pd.read_csv(file_path, sep="\t")
+
+    else:
+        raise ValueError("I don't know how you ended up here")
+
+
+def write_df(df, path_no_format, file_format="parquet"):
+    """Read dataframe from disk with choosen file format
+
+    Parameters
+    ----------
+
+    df: pd.DataFrame
+        Dataframe to save to disk
+
+    path_no_format: str
+        Path for file without format
+
+    file_format: str, default = 'parquet'
+        File format for loading the file. Available options: ['parquet', 'tsv']
+
+    """
+
+    if file_format not in supported_formats:
+        raise ValueError(
+            f"Provided unknown file format: {file_format}, supported_formats: {supported_formats}"
+        )
+
+    file_path = f"{path_no_format}.{file_format}"
+
+    logger.info(f"Saving {file_path} to disk")
+
+    if file_format == "parquet":
+        df.to_parquet(file_path, index=False)
+
+    elif file_format == "tsv":
+        df.to_csv(file_path, sep="\t", index=False)
+
+    else:
+        raise ValueError("I don't know how you ended up here")
diff --git a/alphadia/constants/default.yaml b/alphadia/constants/default.yaml
index 5339a3df..4833a500 100644
--- a/alphadia/constants/default.yaml
+++ b/alphadia/constants/default.yaml
@@ -129,6 +129,8 @@ search_output:
   num_samples_quadratic: 50
   min_nonnan: 3
   normalize_lfq: True
+  # can be either "parquet" or "tsv"
+  file_format: "parquet"
 
 # configuration for the optimization manager
 # initial parameters, will nbe optimized
diff --git a/alphadia/outputtransform.py b/alphadia/outputtransform.py
index 9276cff7..c8ce6985 100644
--- a/alphadia/outputtransform.py
+++ b/alphadia/outputtransform.py
@@ -445,20 +445,10 @@ def load_precursor_table(self):
             Precursor table
         """
 
-        if not os.path.exists(
-            os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}.tsv")
-        ):
-            logger.error(
-                f"Can't continue as no {self.PRECURSOR_OUTPUT}.tsv file was found in the output folder: {self.output_folder}"
-            )
-            raise FileNotFoundError(
-                f"Can't continue as no {self.PRECURSOR_OUTPUT}.tsv file was found in the output folder: {self.output_folder}"
-            )
-        logger.info(f"Reading {self.PRECURSOR_OUTPUT}.tsv file")
-        psm_df = pd.read_csv(
-            os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}.tsv"), sep="\t"
+        return read_df(
+            os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}"),
+            file_type=self.config["file_format"],
         )
-        return psm_df
 
     def build_precursor_table(
         self,
diff --git a/alphadia/peakgroup/search.py b/alphadia/peakgroup/search.py
index 0b97d445..9f3f6a42 100644
--- a/alphadia/peakgroup/search.py
+++ b/alphadia/peakgroup/search.py
@@ -1041,7 +1041,7 @@ def assemble_candidates(self, elution_group_container):
                     precursor_flat_lookup
                 ]
 
-        # save features for training if desired.
+        # DEBUG: save features for training if desired.
         if self.feature_path is not None:
             feature_matrix = np.zeros(
                 (len(candidates), len(candidates[0].features)), dtype=np.float32
diff --git a/tests/unit_tests/conftest.py b/tests/unit_tests/conftest.py
index b1a90b82..1976b9d7 100644
--- a/tests/unit_tests/conftest.py
+++ b/tests/unit_tests/conftest.py
@@ -4,6 +4,7 @@
 import pandas as pd
 import numpy as np
 import matplotlib
+import tempfile
 
 matplotlib.use("Agg")
 from matplotlib import pyplot as plt
@@ -184,3 +185,14 @@ def pytest_configure(config):
         pytest.test_data[raw_folder] = raw_files
 
     # important to supress matplotlib output
+
+
+def _random_tempfolder():
+    tempdir = tempfile.gettempdir()
+    # 6 alphanumeric characters
+    random_foldername = "alphadia_" + "".join(
+        np.random.choice(list("abcdefghijklmnopqrstuvwxyz0123456789"), 6)
+    )
+    path = os.path.join(tempdir, random_foldername)
+    os.mkdir(path)
+    return path
diff --git a/tests/unit_tests/test_consensus.py b/tests/unit_tests/test_consensus.py
new file mode 100644
index 00000000..f592e4ee
--- /dev/null
+++ b/tests/unit_tests/test_consensus.py
@@ -0,0 +1,25 @@
+import pytest
+import pandas as pd
+import os
+from conftest import _random_tempfolder
+from alphadia.consensus.utils import read_df, write_df
+
+
+@pytest.mark.parametrize(
+    "format, should_fail",
+    [("tsv", False), ("parquet", False), ("a321", True)],
+)
+def test_read_write(format, should_fail):
+    # given
+    df = pd.DataFrame([{"a": "a", "b": "b"}, {"a": "a", "b": "b"}])
+    path = os.path.join(_random_tempfolder())
+
+    # when
+    if should_fail:
+        with pytest.raises(ValueError):
+            write_df(df, path, file_format=format)
+
+    else:
+        write_df(df, path, file_format=format)
+        _df = read_df(path, file_format=format)
+        assert df.equals(_df)

From 7876db62e3cc68bccb87775e8d1b082f0360ce1f Mon Sep 17 00:00:00 2001
From: GeorgWa <wallmann@biochem.mpg.de>
Date: Mon, 10 Jun 2024 19:37:12 -0400
Subject: [PATCH 3/5] variable output format

---
 alphadia/consensus/utils.py |  2 +-
 alphadia/outputtransform.py | 39 +++++++++++++++++--------------------
 2 files changed, 19 insertions(+), 22 deletions(-)

diff --git a/alphadia/consensus/utils.py b/alphadia/consensus/utils.py
index 71054635..07638fb6 100644
--- a/alphadia/consensus/utils.py
+++ b/alphadia/consensus/utils.py
@@ -78,7 +78,7 @@ def write_df(df, path_no_format, file_format="parquet"):
         df.to_parquet(file_path, index=False)
 
     elif file_format == "tsv":
-        df.to_csv(file_path, sep="\t", index=False)
+        df.to_csv(file_path, sep="\t", index=False, float_format="%.6f")
 
     else:
         raise ValueError("I don't know how you ended up here")
diff --git a/alphadia/outputtransform.py b/alphadia/outputtransform.py
index c8ce6985..9f1746d6 100644
--- a/alphadia/outputtransform.py
+++ b/alphadia/outputtransform.py
@@ -10,7 +10,7 @@
     TransferLearningAccumulator,
     AccumulationBroadcaster,
 )
-
+from alphadia.consensus.utils import read_df, write_df
 
 import pandas as pd
 import numpy as np
@@ -578,11 +578,10 @@ def build_precursor_table(
             psm_df = psm_df[psm_df["decoy"] == 0]
         if save:
             logger.info("Writing precursor output to disk")
-            psm_df.to_csv(
-                os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}.tsv"),
-                sep="\t",
-                index=False,
-                float_format="%.6f",
+            write_df(
+                psm_df,
+                os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}"),
+                file_format=self.config["search_output"]["file_format"],
             )
 
         return psm_df
@@ -630,11 +629,10 @@ def build_stat_df(
 
         if save:
             logger.info("Writing stat output to disk")
-            stat_df.to_csv(
-                os.path.join(self.output_folder, f"{self.STAT_OUTPUT}.tsv"),
-                sep="\t",
-                index=False,
-                float_format="%.6f",
+            write_df(
+                stat_df,
+                os.path.join(self.output_folder, f"{self.STAT_OUTPUT}"),
+                file_format="tsv",
             )
 
         return stat_df
@@ -712,11 +710,11 @@ def build_lfq_tables(
 
             if save:
                 logger.info(f"Writing {group_nice} output to disk")
-                lfq_df.to_csv(
-                    os.path.join(self.output_folder, f"{group_nice}.matrix.tsv"),
-                    sep="\t",
-                    index=False,
-                    float_format="%.6f",
+
+                write_df(
+                    lfq_df,
+                    os.path.join(self.output_folder, f"{group_nice}.matrix"),
+                    file_format=self.config["search_output"]["file_format"],
                 )
 
         protein_df_melted = lfq_df.melt(
@@ -727,11 +725,10 @@ def build_lfq_tables(
 
         if save:
             logger.info("Writing psm output to disk")
-            psm_df.to_csv(
-                os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}.tsv"),
-                sep="\t",
-                index=False,
-                float_format="%.6f",
+            write_df(
+                psm_df,
+                os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}"),
+                file_format=self.config["search_output"]["file_format"],
             )
 
         return lfq_df

From 74bb3621a3817eba07200843dbb76b77a3669f61 Mon Sep 17 00:00:00 2001
From: GeorgWa <wallmann@biochem.mpg.de>
Date: Mon, 10 Jun 2024 19:58:58 -0400
Subject: [PATCH 4/5] fix tests

---
 alphadia/outputtransform.py                |  2 +-
 tests/unit_tests/conftest.py               |  4 ++--
 tests/unit_tests/test_consensus.py         |  4 ++--
 tests/unit_tests/test_outputaccumulator.py |  6 +++---
 tests/unit_tests/test_outputtransform.py   | 11 ++++++-----
 tests/unit_tests/test_reporting.py         | 23 +++++++---------------
 6 files changed, 21 insertions(+), 29 deletions(-)

diff --git a/alphadia/outputtransform.py b/alphadia/outputtransform.py
index 9f1746d6..75a0db4b 100644
--- a/alphadia/outputtransform.py
+++ b/alphadia/outputtransform.py
@@ -447,7 +447,7 @@ def load_precursor_table(self):
 
         return read_df(
             os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}"),
-            file_type=self.config["file_format"],
+            file_format=self.config["search_output"]["file_format"],
         )
 
     def build_precursor_table(
diff --git a/tests/unit_tests/conftest.py b/tests/unit_tests/conftest.py
index 1976b9d7..5f977eb9 100644
--- a/tests/unit_tests/conftest.py
+++ b/tests/unit_tests/conftest.py
@@ -187,12 +187,12 @@ def pytest_configure(config):
     # important to supress matplotlib output
 
 
-def _random_tempfolder():
+def random_tempfolder():
     tempdir = tempfile.gettempdir()
     # 6 alphanumeric characters
     random_foldername = "alphadia_" + "".join(
         np.random.choice(list("abcdefghijklmnopqrstuvwxyz0123456789"), 6)
     )
     path = os.path.join(tempdir, random_foldername)
-    os.mkdir(path)
+    os.makedirs(path, exist_ok=True)
     return path
diff --git a/tests/unit_tests/test_consensus.py b/tests/unit_tests/test_consensus.py
index f592e4ee..65aa5b19 100644
--- a/tests/unit_tests/test_consensus.py
+++ b/tests/unit_tests/test_consensus.py
@@ -1,7 +1,7 @@
 import pytest
 import pandas as pd
 import os
-from conftest import _random_tempfolder
+from conftest import random_tempfolder
 from alphadia.consensus.utils import read_df, write_df
 
 
@@ -12,7 +12,7 @@
 def test_read_write(format, should_fail):
     # given
     df = pd.DataFrame([{"a": "a", "b": "b"}, {"a": "a", "b": "b"}])
-    path = os.path.join(_random_tempfolder())
+    path = os.path.join(random_tempfolder())
 
     # when
     if should_fail:
diff --git a/tests/unit_tests/test_outputaccumulator.py b/tests/unit_tests/test_outputaccumulator.py
index 097e59f5..9de649b4 100644
--- a/tests/unit_tests/test_outputaccumulator.py
+++ b/tests/unit_tests/test_outputaccumulator.py
@@ -98,9 +98,9 @@ def prepare_input_data():
 
     for i, raw_folder in enumerate(raw_folders):
         os.makedirs(raw_folder, exist_ok=True)
-        psm_dfs[i].to_csv(os.path.join(raw_folder, "psm.tsv"), sep="\t", index=False)
-        fragment_dfs[i].to_csv(
-            os.path.join(raw_folder, "frag.tsv"), sep="\t", index=False
+        psm_dfs[i].to_parquet(os.path.join(raw_folder, "psm.parquet"), index=False)
+        fragment_dfs[i].to_parquet(
+            os.path.join(raw_folder, "frag.parquet"), index=False
         )
 
     return config, temp_folder, raw_folders, psm_dfs, fragment_dfs
diff --git a/tests/unit_tests/test_outputtransform.py b/tests/unit_tests/test_outputtransform.py
index 91a18926..bdbe038a 100644
--- a/tests/unit_tests/test_outputtransform.py
+++ b/tests/unit_tests/test_outputtransform.py
@@ -28,6 +28,7 @@ def test_output_transform():
             "normalize_lfq": True,
             "peptide_level_lfq": False,
             "precursor_level_lfq": False,
+            "file_format": "parquet",
         },
     }
 
@@ -52,8 +53,8 @@ def test_output_transform():
             fragment_base_df["precursor_idx"].isin(psm_df["precursor_idx"])
         ]
 
-        frag_df.to_csv(os.path.join(raw_folder, "frag.tsv"), sep="\t", index=False)
-        psm_df.to_csv(os.path.join(raw_folder, "psm.tsv"), sep="\t", index=False)
+        frag_df.to_parquet(os.path.join(raw_folder, "frag.parquet"), index=False)
+        psm_df.to_parquet(os.path.join(raw_folder, "psm.parquet"), index=False)
 
     output = outputtransform.SearchPlanOutput(config, temp_folder)
     _ = output.build_precursor_table(raw_folders, save=True)
@@ -61,8 +62,8 @@ def test_output_transform():
     _ = output.build_lfq_tables(raw_folders, save=True)
 
     # validate psm_df output
-    psm_df = pd.read_csv(
-        os.path.join(temp_folder, f"{output.PRECURSOR_OUTPUT}.tsv"), sep="\t"
+    psm_df = pd.read_parquet(
+        os.path.join(temp_folder, f"{output.PRECURSOR_OUTPUT}.parquet"),
     )
     assert all(
         [
@@ -91,7 +92,7 @@ def test_output_transform():
     assert all([col in stat_df.columns for col in ["run", "precursors", "proteins"]])
 
     # validate protein_df output
-    protein_df = pd.read_csv(os.path.join(temp_folder, "pg.matrix.tsv"), sep="\t")
+    protein_df = pd.read_parquet(os.path.join(temp_folder, "pg.matrix.parquet"))
     assert all([col in protein_df.columns for col in ["run_0", "run_1", "run_2"]])
 
     for i in run_columns:
diff --git a/tests/unit_tests/test_reporting.py b/tests/unit_tests/test_reporting.py
index ed536de0..5d171f74 100644
--- a/tests/unit_tests/test_reporting.py
+++ b/tests/unit_tests/test_reporting.py
@@ -7,23 +7,14 @@
 import sys
 import pytest
 
-from alphadia.workflow import reporting
-
+from conftest import random_tempfolder
 
-def _random_tempfolder():
-    tempdir = tempfile.gettempdir()
-    # 6 alphanumeric characters
-    random_foldername = "".join(
-        np.random.choice(list("abcdefghijklmnopqrstuvwxyz0123456789"), 6)
-    )
-    path = os.path.join(tempdir, random_foldername)
-    os.mkdir(path)
-    return path
+from alphadia.workflow import reporting
 
 
 @pytest.mark.skipif(sys.platform == "win32", reason="does not run on windows")
 def test_logging():
-    tempfolder = _random_tempfolder()
+    tempfolder = random_tempfolder()
 
     if os.path.exists(os.path.join(tempfolder, "log.txt")):
         os.remove(os.path.join(tempfolder, "log.txt"))
@@ -58,7 +49,7 @@ def test_backend():
 
 
 def test_figure_backend():
-    tempfolder = _random_tempfolder()
+    tempfolder = random_tempfolder()
 
     figure_backend = reporting.FigureBackend(path=tempfolder)
 
@@ -79,7 +70,7 @@ def test_figure_backend():
 
 
 def test_jsonl_backend():
-    tempfolder = _random_tempfolder()
+    tempfolder = random_tempfolder()
 
     with reporting.JSONLBackend(path=tempfolder) as jsonl_backend:
         jsonl_backend.log_event("start_extraction", None)
@@ -96,7 +87,7 @@ def test_jsonl_backend():
 
 @pytest.mark.skipif(sys.platform == "win32", reason="does not run on windows")
 def test_log_backend():
-    tempfolder = _random_tempfolder()
+    tempfolder = random_tempfolder()
 
     if os.path.exists(os.path.join(tempfolder, "log.txt")):
         os.remove(os.path.join(tempfolder, "log.txt"))
@@ -117,7 +108,7 @@ def test_log_backend():
 
 @pytest.mark.skipif(sys.platform == "win32", reason="does not run on windows")
 def test_pipeline():
-    tempfolder = _random_tempfolder()
+    tempfolder = random_tempfolder()
 
     pipeline = reporting.Pipeline(
         backends=[

From db1e3f4c8cae05085f52ac9c43e0f6818c8b9025 Mon Sep 17 00:00:00 2001
From: GeorgWa <wallmann@biochem.mpg.de>
Date: Wed, 12 Jun 2024 18:03:53 -0700
Subject: [PATCH 5/5] implement fixes

---
 alphadia/consensus/utils.py     |  9 +++------
 alphadia/constants/default.yaml |  2 +-
 alphadia/outputtransform.py     |  4 ++--
 tests/unit_tests/conftest.py    | 12 ++++++++++++
 4 files changed, 18 insertions(+), 9 deletions(-)

diff --git a/alphadia/consensus/utils.py b/alphadia/consensus/utils.py
index 07638fb6..084ded8d 100644
--- a/alphadia/consensus/utils.py
+++ b/alphadia/consensus/utils.py
@@ -26,11 +26,6 @@ def read_df(path_no_format, file_format="parquet"):
 
     """
 
-    if file_format not in supported_formats:
-        raise ValueError(
-            f"Provided unknown file format: {file_format}, supported_formats: {supported_formats}"
-        )
-
     file_path = f"{path_no_format}.{file_format}"
 
     if not os.path.exists(file_path):
@@ -45,7 +40,9 @@ def read_df(path_no_format, file_format="parquet"):
         return pd.read_csv(file_path, sep="\t")
 
     else:
-        raise ValueError("I don't know how you ended up here")
+        raise ValueError(
+            f"Provided unknown file format: {file_format}, supported_formats: {supported_formats}"
+        )
 
 
 def write_df(df, path_no_format, file_format="parquet"):
diff --git a/alphadia/constants/default.yaml b/alphadia/constants/default.yaml
index 4833a500..3d3a8112 100644
--- a/alphadia/constants/default.yaml
+++ b/alphadia/constants/default.yaml
@@ -130,7 +130,7 @@ search_output:
   min_nonnan: 3
   normalize_lfq: True
   # can be either "parquet" or "tsv"
-  file_format: "parquet"
+  file_format: "tsv"
 
 # configuration for the optimization manager
 # initial parameters, will nbe optimized
diff --git a/alphadia/outputtransform.py b/alphadia/outputtransform.py
index 75a0db4b..f5db4491 100644
--- a/alphadia/outputtransform.py
+++ b/alphadia/outputtransform.py
@@ -580,7 +580,7 @@ def build_precursor_table(
             logger.info("Writing precursor output to disk")
             write_df(
                 psm_df,
-                os.path.join(self.output_folder, f"{self.PRECURSOR_OUTPUT}"),
+                os.path.join(self.output_folder, self.PRECURSOR_OUTPUT),
                 file_format=self.config["search_output"]["file_format"],
             )
 
@@ -631,7 +631,7 @@ def build_stat_df(
             logger.info("Writing stat output to disk")
             write_df(
                 stat_df,
-                os.path.join(self.output_folder, f"{self.STAT_OUTPUT}"),
+                os.path.join(self.output_folder, self.STAT_OUTPUT),
                 file_format="tsv",
             )
 
diff --git a/tests/unit_tests/conftest.py b/tests/unit_tests/conftest.py
index 5f977eb9..3ddd2017 100644
--- a/tests/unit_tests/conftest.py
+++ b/tests/unit_tests/conftest.py
@@ -24,6 +24,9 @@ def mock_precursor_df(
     n_precursor : int
         Number of precursors to generate
 
+    with_decoy : bool
+        If True, half of the precursors will be decoys
+
     Returns
     -------
 
@@ -188,6 +191,14 @@ def pytest_configure(config):
 
 
 def random_tempfolder():
+    """Create a randomly named temp folder in the system temp folder
+
+    Returns
+    -------
+    path : str
+        Path to the created temp folder
+
+    """
     tempdir = tempfile.gettempdir()
     # 6 alphanumeric characters
     random_foldername = "alphadia_" + "".join(
@@ -195,4 +206,5 @@ def random_tempfolder():
     )
     path = os.path.join(tempdir, random_foldername)
     os.makedirs(path, exist_ok=True)
+    print(f"Created temp folder: {path}")
     return path