fix float output format for consistency

metatlas/io/metatlas_get_data_helper_fun.py

@@ -771,15 +771,15 @@ def make_data_sources_tables(groups, myatlas, output_loc, polarity=None, overwri
     output_dir = os.path.join(output_loc, f"{prefix}data_sources")
     atlas_path = os.path.join(output_dir, f"{prefix}atlas_metadata.tab")
     write_utils.export_dataframe(metob.to_dataframe([myatlas]), atlas_path, "atlas metadata",
-                                 overwrite, sep='\t')
+                                 overwrite, sep='\t', float_format="%.8e")
     groups_path = os.path.join(output_dir, f"{prefix}groups_metadata.tab")
     write_utils.export_dataframe(metob.to_dataframe(groups), groups_path, "groups metadata",
                                  overwrite, sep='\t')
 
     atlas_df = make_atlas_df(myatlas)
     atlas_df['label'] = [cid.name for cid in myatlas.compound_identifications]
     atlas_df_path = os.path.join(output_dir, myatlas.name+'_originalatlas.tab')
-    write_utils.export_dataframe(atlas_df, atlas_df_path, "atlas dataframe", overwrite, sep='\t')
+    write_utils.export_dataframe(atlas_df, atlas_df_path, "atlas dataframe", overwrite, sep='\t', float_format="%.6e")
 
     group_path_df = pd.DataFrame(columns=['group_name', 'group_path', 'file_name'])
     loc_counter = 0

metatlas/io/targeted_output.py

@@ -26,7 +26,7 @@ def write_atlas_to_spreadsheet(metatlas_dataset, overwrite=False):
     """Save atlas as csv file. Will not overwrite existing file unless overwrite is True"""
     out_file_name = os.path.join(metatlas_dataset.ids.output_dir, f"{metatlas_dataset.atlas.name}_export.csv")
     out_df = dp.export_atlas_to_spreadsheet(metatlas_dataset.atlas)
-    write_utils.export_dataframe_die_on_diff(out_df, out_file_name, "atlas", overwrite=overwrite)
+    write_utils.export_dataframe_die_on_diff(out_df, out_file_name, "atlas", overwrite=overwrite, float_format="%.6e")
 
 
 def write_stats_table(
@@ -75,7 +75,7 @@ def write_stats_table(
         min_num_frag_matches,
         min_relative_frag_intensity,
     )
-    write_utils.export_dataframe_die_on_diff(scores_df, scores_path, "scores", overwrite=overwrite)
+    write_utils.export_dataframe_die_on_diff(scores_df, scores_path, "scores", overwrite=overwrite, float_format="%.8e")
     fa.make_stats_table(
         input_dataset=metatlas_dataset,
         msms_hits=metatlas_dataset.hits,
@@ -204,7 +204,7 @@ def write_msms_fragment_ions(
         )
     out_df = pd.DataFrame(out)
     path = os.path.join(data.ids.output_dir, f"spectra_{intensity_fraction:.2f}pct_{int(min_mz)}cut.csv")
-    write_utils.export_dataframe_die_on_diff(out_df, path, "MSMS fragment ions", overwrite=overwrite)
+    write_utils.export_dataframe_die_on_diff(out_df, path, "MSMS fragment ions", overwrite=overwrite, float_format="%.8e")
     return out_df
 
 

metatlas/plots/dill2plots.py

@@ -1464,7 +1464,7 @@ def make_output_dataframe(input_fname='', input_dataset=None, include_lcmsruns=N
     if output_loc:
         prefix = f"{polarity}_" if polarity != '' else ''
         df_path = os.path.join(output_loc, f"{prefix}{fieldname}.tab")
-        write_utils.export_dataframe_die_on_diff(out, df_path, fieldname, overwrite=overwrite, sep="\t")
+        write_utils.export_dataframe_die_on_diff(out, df_path, fieldname, overwrite=overwrite, sep="\t", float_format="%.9e")
     return out
 
 
@@ -2361,7 +2361,7 @@ def no_axis_plot(i):
         plt.close()
         logger.debug('Exported identification figures for %s to %s.', compound_names[compound_idx], fig_path)
     match_path = os.path.join(output_loc, 'MatchingMZs.tab')
-    write_utils.export_dataframe(match, match_path, 'matching MZs', overwrite, sep='\t')
+    write_utils.export_dataframe(match, match_path, 'matching MZs', overwrite, sep='\t', float_format="%.12e")
 
 
 def plot_ms1_spectra(polarity = None, mz_min = 5, mz_max = 5, input_fname = '', input_dataset = [], compound_names = [],  include_lcmsruns = [], exclude_lcmsruns = [], include_groups = [], exclude_groups = [], output_loc = []):

metatlas/tools/fastanalysis.py

@@ -409,7 +409,7 @@ def make_stats_table(input_fname = '', input_dataset = [], msms_hits_df = None,
     if output_loc is not None:
         stats_tables_dir = os.path.join(output_loc, f"{prefix}stats_tables")
         stats_path = os.path.join(stats_tables_dir, f"{prefix}stats_table.tab")
-        write_utils.export_dataframe_die_on_diff(stats_table, stats_path, 'stats table', overwrite, sep='\t')
+        write_utils.export_dataframe_die_on_diff(stats_table, stats_path, 'stats table', overwrite, sep='\t', float_format="%.8e")
         readme_path = os.path.join(stats_tables_dir, f"{prefix}stats_table.readme")
         write_utils.check_existing_file(readme_path, overwrite)
         with open(readme_path, 'w') as readme:

metatlas/tools/predict_rt.py

@@ -240,13 +240,13 @@ def load_runs(files_df, qc_atlas_df, qc_atlas, cpus):
 def save_measured_rts(metatlas_dataset, file_name):
     """Save RT values in csv format file"""
     rts_df = get_rts(metatlas_dataset, include_atlas_rt_peak=False)
-    write_utils.export_dataframe_die_on_diff(rts_df, file_name, "measured RT values")
+    write_utils.export_dataframe_die_on_diff(rts_df, file_name, "measured RT values", float_format="%.6e")
 
 
 def save_rt_peak(metatlas_dataset, file_name):
     """Save peak RT values in tsv format file"""
     rts_df = dp.make_output_dataframe(input_dataset=metatlas_dataset, fieldname="rt_peak", use_labels=True)
-    write_utils.export_dataframe_die_on_diff(rts_df, file_name, "peak RT values", sep="\t")
+    write_utils.export_dataframe_die_on_diff(rts_df, file_name, "peak RT values", sep="\t", float_format="%.6e")
 
 
 def get_rts(metatlas_dataset, include_atlas_rt_peak=True):
@@ -426,7 +426,7 @@ def save_model_comparison(selected_column, qc_atlas_df, rts_df, linear, poly, fi
     # qc_df["RT Polynomial Pred"] = poly.predict(qc_df["RT Reference"].to_numpy())
     qc_df["RT Diff Linear"] = qc_df["RT Measured"] - qc_df["RT Linear Pred"]
     qc_df["RT Diff Polynomial"] = qc_df["RT Measured"] - qc_df["RT Polynomial Pred"]
-    write_utils.export_dataframe_die_on_diff(qc_df, file_name, "model comparision")
+    write_utils.export_dataframe_die_on_diff(qc_df, file_name, "model comparision", float_format="%.6e")
 
 
 def write_models(file_name, linear_model, poly_model, groups, atlas):
@@ -501,7 +501,7 @@ def create_adjusted_atlases(linear, poly, ids, atlas_indices=None, free_text="",
         prd_atlas_df["rt_min"] = prd_atlas_df["rt_peak"].apply(lambda rt: rt - 0.5)
         prd_atlas_df["rt_max"] = prd_atlas_df["rt_peak"].apply(lambda rt: rt + 0.5)
         write_utils.export_dataframe_die_on_diff(
-            prd_atlas_df, prd_atlas_file_name, "predicted atlas", index=False
+            prd_atlas_df, prd_atlas_file_name, "predicted atlas", index=False, float_format="%.6e"
         )
         if save_to_db:
             dp.make_atlas_from_spreadsheet(