masster 0.4.20__py3-none-any.whl → 0.4.21__py3-none-any.whl

masster/__init__.py

@@ -8,6 +8,12 @@ mass spectrometry workflows.
 
 from __future__ import annotations
 
+import warnings
+
+# Suppress pyOpenMS environment variable warnings globally
+warnings.filterwarnings("ignore", message=".*OPENMS_DATA_PATH.*", category=UserWarning)
+warnings.filterwarnings("ignore", message="Warning: OPENMS_DATA_PATH.*", category=UserWarning)
+
 from masster._version import __version__
 
 # from masster._version import get_version

masster/_version.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 
-__version__ = "0.4.20"
+__version__ = "0.4.21"
 
 
 def get_version():

masster/sample/h5.py

@@ -235,6 +235,22 @@ def _save_sample5(
                             data=serialized_data,
                             compression="gzip",
                         )
+                    elif col == "ms1_spec":
+                        # this column contains either None or numpy arrays with isotope pattern data
+                        # serialize numpy arrays to JSON strings for storage
+                        data = features[col]
+                        data_as_json_strings = []
+                        for i in range(len(data)):
+                            if data[i] is not None:
+                                # Convert numpy array to list and then to JSON
+                                data_as_json_strings.append(json.dumps(data[i].tolist()))
+                            else:
+                                data_as_json_strings.append("None")
+                        features_group.create_dataset(
+                            col,
+                            data=data_as_json_strings,
+                            compression="gzip",
+                        )
 
                     else:
                         self.logger.warning(
@@ -630,6 +646,25 @@ def _load_sample5(self, filename: str, map: bool = False):
                                                 )
                                         reconstructed_data.append(spectrum_list)
 
+                                data[col] = reconstructed_data
+                            case "ms1_spec":
+                                data_col = features_group[col][:]
+                                # Convert JSON strings back to numpy arrays
+                                reconstructed_data = []
+                                for item in data_col:
+                                    if isinstance(item, bytes):
+                                        item = item.decode("utf-8")
+
+                                    if item == "None" or item == "":
+                                        reconstructed_data.append(None)
+                                    else:
+                                        try:
+                                            # Parse JSON string to get list and convert to numpy array
+                                            array_data = json.loads(item)
+                                            reconstructed_data.append(np.array(array_data, dtype=np.float64))
+                                        except (json.JSONDecodeError, ValueError, TypeError):
+                                            reconstructed_data.append(None)
+
                                 data[col] = reconstructed_data
                             case _:
                                 self.logger.debug(f"Unexpected Object column '{col}'")
@@ -1371,6 +1406,25 @@ def _load_sample5_study(self, filename: str, map: bool = False):
                                         ):
                                             reconstructed_data.append(None)
 
+                                data[col] = reconstructed_data
+                            case "ms1_spec":
+                                data_col = features_group[col][:]
+                                # Convert JSON strings back to numpy arrays
+                                reconstructed_data = []
+                                for item in data_col:
+                                    if isinstance(item, bytes):
+                                        item = item.decode("utf-8")
+
+                                    if item == "None" or item == "":
+                                        reconstructed_data.append(None)
+                                    else:
+                                        try:
+                                            # Parse JSON string to get list and convert to numpy array
+                                            array_data = json.loads(item)
+                                            reconstructed_data.append(np.array(array_data, dtype=np.float64))
+                                        except (json.JSONDecodeError, ValueError, TypeError):
+                                            reconstructed_data.append(None)
+
                                 data[col] = reconstructed_data
                             case _:
                                 # Handle other Object columns as raw data
@@ -1407,6 +1461,9 @@ def _load_sample5_study(self, filename: str, map: bool = False):
                 # Add Object columns one by one
                 for col, values in object_columns.items():
                     if not self.features_df.is_empty():
+                        # Fix for missing columns: if values is None, create list of None with correct length
+                        if values is None:
+                            values = [None] * len(self.features_df)
                         self.features_df = self.features_df.with_columns(
                             pl.Series(col, values, dtype=pl.Object).alias(col),
                         )
@@ -2027,7 +2084,7 @@ def load_dataframe_from_h5_group(
     for col in schema_columns:
         if col not in group:
             if logger:
-                logger.warning(f"Column '{col}' not found in {df_name}.")
+                logger.info(f"Column '{col}' not found in {df_name}.")
             data[col] = None
             missing_columns.append(col)
             continue

masster/sample/load.py

@@ -48,9 +48,14 @@ from tqdm import tqdm
 from masster.chromatogram import Chromatogram
 from masster.spectrum import Spectrum
 
+# Suppress pyOpenMS warnings globally
+warnings.filterwarnings("ignore", message=".*OPENMS_DATA_PATH.*", category=UserWarning)
+warnings.filterwarnings("ignore", message="Warning: OPENMS_DATA_PATH.*", category=UserWarning)
+
 # Import pyopenms with suppressed warnings
 with warnings.catch_warnings():
-    warnings.filterwarnings("ignore", message="Warning: OPENMS_DATA_PATH environment variable already exists.*", category=UserWarning)
+    warnings.filterwarnings("ignore", message=".*OPENMS_DATA_PATH environment variable already exists.*", category=UserWarning)
+    warnings.filterwarnings("ignore", message="Warning: OPENMS_DATA_PATH.*", category=UserWarning)
     import pyopenms as oms
 
 
@@ -633,6 +638,7 @@ def _load_wiff(
             mz=peaks.mz.values,
             inty=peaks.intensity.values,
             ms_level=ms_level,
+            centroided=False,  # WIFF files always contain profile data
         )
         bl = spect.baseline()
         spect = spect.denoise(threshold=bl)

masster/sample/plot.py

@@ -387,18 +387,19 @@ def plot_2d(
     show_only_features_with_ms2=False,
     show_isotopes=False,
     show_ms2=False,
+    show_in_browser=False,
     title=None,
     cmap=None,
     marker="circle",
     markersize=10,
-    size="dynamic",
+    size="static",
     raster_dynamic=True,
     raster_max_px=8,
     raster_threshold=0.8,
     height=600,
     width=800,
     mz_range=None,
-    rt_range=None,
+    rt_range=None
 ):
     """
     Plot a two-dimensional visualization of MS1 survey scan data with optional overlays
@@ -634,8 +635,10 @@ def plot_2d(
                 ("m/z", "@mz{0.0000}"),
                 ("feature_uid", "@feature_uid"),
                 ("inty", "@inty"),
-                ("quality", "@quality"),
-                ("rt_delta", "@rt_delta"),
+                ("iso", "@iso"),
+                ("adduct", "@adduct"),
+                ("chrom_coherence", "@chrom_coherence"),
+                ("chrom_prominence_scaled", "@chrom_prominence_scaled"),
             ],
         )
         feature_points_1 = hv.Points(
@@ -644,8 +647,8 @@ def plot_2d(
             vdims=[
                 "feature_uid",
                 "inty",
-                "quality",
-                "rt_delta",
+                "iso",
+                "adduct",
                 "ms2_scans",
                 "chrom_coherence",
                 "chrom_prominence_scaled",
@@ -666,8 +669,10 @@ def plot_2d(
                 ("m/z", "@mz{0.0000}"),
                 ("feature_uid", "@feature_uid"),
                 ("inty", "@inty"),
-                ("quality", "@quality"),
-                ("rt_delta", "@rt_delta"),
+                ("iso", "@iso"),
+                ("adduct", "@adduct"),
+                ("chrom_coherence", "@chrom_coherence"),
+                ("chrom_prominence_scaled", "@chrom_prominence_scaled"),
             ],
         )
         feature_points_2 = hv.Points(
@@ -676,8 +681,8 @@ def plot_2d(
             vdims=[
                 "feature_uid",
                 "inty",
-                "quality",
-                "rt_delta",
+                "iso",
+                "adduct",
                 "chrom_coherence",
                 "chrom_prominence_scaled",
             ],
@@ -702,10 +707,11 @@ def plot_2d(
                     ("m/z", "@mz{0.0000}"),
                     ("feature_uid", "@feature_uid"),
                     ("inty", "@inty"),
-                    ("quality", "@quality"),
-                    ("rt_delta", "@rt_delta"),
                     ("iso", "@iso"),
                     ("iso_of", "@iso_of"),
+                    ("adduct", "@adduct"),
+                    ("chrom_coherence", "@chrom_coherence"),
+                    ("chrom_prominence_scaled", "@chrom_prominence_scaled"),
                 ],
             )
             feature_points_iso = hv.Points(
@@ -714,10 +720,9 @@ def plot_2d(
                 vdims=[
                     "feature_uid",
                     "inty",
-                    "quality",
-                    "rt_delta",
                     "iso",
                     "iso_of",
+                    "adduct",
                     "chrom_coherence",
                     "chrom_prominence_scaled",
                 ],
@@ -918,21 +923,24 @@ def plot_2d(
             else:
                 # For slider plots, save the current state
                 hv.save(create_feature_overlay(markersize), filename, fmt="png")
-            return None
         else:
-            # For notebook display, return the interactive layout
-            return _display_plot(layout, layout)
+            # Use show() for display in notebook
+            layout.show()
     else:
         # Create a panel layout without slider
         layout = panel.Column(overlay)
 
+    # Handle display logic based on show_in_browser and raster_dynamic
     if filename is not None:
         # Use consistent save/display behavior
         self._handle_sample_plot_output(layout, filename, "panel")
-        return None
     else:
-        # Check if we're in a notebook environment and display appropriately
-        return _display_plot(overlay, layout)
+        # Show in browser if both show_in_browser and raster_dynamic are True
+        if show_in_browser and raster_dynamic:
+            layout.show()
+        else:
+            # Return to notebook for inline display
+            return layout
 
 
 def plot_2d_oracle(
@@ -1952,11 +1960,10 @@ def plot_feature_stats(
     filename=None,
 ):
     """
-    Generates overlaid distribution plots for selected feature metrics.
+    Generates vertically stacked density plots for selected feature metrics.
     The distributions are created separately for features with and without MS2 data.
-    Metrics include intensity, quality, retention time, m/z (and m/z delta), number of MS2 peaks,
-    summed MS2 intensities, and the MS2-to-MS1 ratio. The plots help to visualize the distribution
-    differences between features that are linked to MS2 spectra and those that are not.
+    Metrics include mz, rt, log10(inty), chrom_coherence, chrom_prominence, and chrom_prominence_scaled.
+    The plots help to visualize the distribution differences between features that are linked to MS2 spectra and those that are not.
 
     Parameters:
         filename (str, optional): The output filename. If the filename ends with ".html",
@@ -1972,54 +1979,28 @@ def plot_feature_stats(
     # Convert to pandas for operations that require pandas functionality
     if hasattr(feats, "to_pandas"):
         feats = feats.to_pandas()
-    # Compute m/z delta for each feature
-    feats["mz_delta"] = feats["mz_end"] - feats["mz_start"]
-    # Add a column with the number of peaks in the MS2 spectrum
-    feats["MS2peaks"] = feats["ms2_specs"].apply(
-        lambda x: len(x[0]) if x is not None else 0,
-    )
-    # Add a column with the sum of intensities in the MS2 spectrum
-    feats["MS2int"] = feats["ms2_specs"].apply(
-        lambda x: sum(x[0].inty) if x is not None else 0,
-    )
 
-    # Calculate the ratio of MS2 to MS1 intensities
-    feats["MS2toMS1"] = feats["MS2int"] / feats["inty"]
-    # Apply log10 transformation to intensity, quality, and MS2int columns (handling non-positive values)
+    # Apply log10 transformation to intensity (handling non-positive values)
     feats["inty"] = np.where(feats["inty"] <= 0, np.nan, np.log10(feats["inty"]))
-    # COMMENT: AR was bugging
-    # feats["chrom_heights"] = np.where(
-    #     feats["chrom_heights"] <= 0, np.nan, np.log10(feats["chrom_heights"])
-    # )
-
-    feats["quality"] = np.where(
-        feats["quality"] <= 0,
-        np.nan,
-        np.log10(feats["quality"]),
-    )
-    feats["MS2int"] = np.where(feats["MS2int"] <= 0, np.nan, np.log10(feats["MS2int"]))
+    
+    # Apply log10 transformation to quality (handling non-positive values)
+    feats["quality"] = np.where(feats["quality"] <= 0, np.nan, np.log10(feats["quality"]))
 
     # Separate features based on presence of MS2 data
     feats_with_MS2 = feats[feats["ms2_scans"].notnull()]
     feats_without_MS2 = feats[feats["ms2_scans"].isnull()]
 
-    # Define the metrics to plot
+    # Define the specific metrics to plot
     cols_to_plot = [
         "mz",
-        "mz_delta",
-        "inty",
-        "quality",
-        "rt",
+        "rt", 
+        "inty",  # Already log10 transformed above
         "rt_delta",
+        "quality",  # Already log10 transformed above
         "chrom_coherence",
         "chrom_prominence",
         "chrom_prominence_scaled",
-        # COMMENT: AR was bugging
-        # "chrom_heights",
-        # "chrom_heights_scaled",
-        "MS2peaks",
-        "MS2int",
-        "MS2toMS1",
+        "chrom_height_scaled",
     ]
 
     # Ensure an index column is available for plotting
@@ -2032,29 +2013,39 @@ def plot_feature_stats(
         data_with = feats_with_MS2[col].dropna().values
         data_without = feats_without_MS2[col].dropna().values
 
-        # Create distribution elements for features with and without MS2
+        # Create distribution elements - Green for WITH MS2, Red for WITHOUT MS2
         dist_with = hv.Distribution(data_with, label="With MS2").opts(
-            color="red",
+            color="green",
             alpha=0.6,
         )
         dist_without = hv.Distribution(data_without, label="Without MS2").opts(
-            color="blue",
+            color="red", 
             alpha=0.6,
         )
 
         # Overlay the distributions with a legend and hover tool enabled
+        title = col
+        if col == "inty":
+            title = "log10(inty)"
+        elif col == "quality":
+            title = "log10(quality)"
+            
         overlay = (dist_with * dist_without).opts(
-            title=col,
+            title=title,
             show_legend=True,
             tools=["hover"],
         )
         density_plots.append(overlay)
 
-    # Arrange the plots in a layout with three columns
+    # Arrange the plots in a grid layout (3 columns for 7 plots)
     layout = hv.Layout(density_plots).cols(3).opts(shared_axes=False)
 
     # Use consistent save/display behavior
-    self._handle_sample_plot_output(layout, filename, "holoviews")
+    if filename is not None:
+        self._handle_sample_plot_output(layout, filename, "holoviews")
+    else:
+        # Return the layout directly for notebook display
+        return layout
 
 
 def plot_tic(

masster/sample/processing.py

@@ -1273,3 +1273,161 @@ def find_ms2(self, **kwargs):
     self.logger.debug(
         "Parameters stored to find_ms2",
     )
+
+
+def find_iso(self, rt_tolerance: float = 0.1, **kwargs):
+    """Extract isotopic distributions from MS1 data and add to features_df.
+    
+    This method processes each feature to find isotopic distributions from MS1 data,
+    similar to the study.find_iso() method but for individual samples. The method
+    adds a new 'ms1_spec' column to features_df containing numpy arrays with
+    isotopic distribution data.
+    
+    Args:
+        rt_tolerance (float): RT tolerance in minutes for matching MS1 scans. Default 0.1.
+        **kwargs: Additional parameters
+    
+    Notes:
+        - Adds a new 'ms1_spec' column to features_df containing numpy arrays
+        - Each array contains [mz, intensity] pairs for the isotopic distribution
+        - Uses the same isotope shift pattern as study.find_iso()
+        - Only processes features that don't already have ms1_spec data
+    """
+    if self.features_df is None or self.features_df.is_empty():
+        self.logger.warning("No features found. Run find_features() first.")
+        return
+        
+    if self.ms1_df is None or self.ms1_df.is_empty():
+        self.logger.warning("No MS1 data found.")
+        return
+    
+    # Check if ms1_spec column already exists
+    if "ms1_spec" in self.features_df.columns:
+        features_without_spec = self.features_df.filter(pl.col("ms1_spec").is_null())
+        if features_without_spec.is_empty():
+            self.logger.info("All features already have isotopic distributions.")
+            return
+        self.logger.info(f"Processing {len(features_without_spec)} features without isotopic distributions.")
+    else:
+        # Add the ms1_spec column with None values
+        self.features_df = self.features_df.with_columns(
+            pl.lit(None, dtype=pl.Object).alias("ms1_spec")
+        )
+        features_without_spec = self.features_df
+        self.logger.info(f"Processing {len(features_without_spec)} features for isotopic distributions.")
+
+    # Define isotope shifts (same as study.find_iso)
+    isotope_shifts = np.array([
+        0.33,
+        0.50,
+        0.66,
+        1.00335,
+        1.50502,
+        2.00670,
+        3.01005,
+        4.01340,
+        5.01675,
+        6.02010,
+        7.02345,
+    ])
+    
+    # Convert rt_tolerance from minutes to seconds
+    rt_tolerance_s = rt_tolerance * 60
+    
+    # Process each feature
+    ms1_specs = []
+    feature_indices = []
+    
+    for i, row in enumerate(tqdm(
+        features_without_spec.rows(named=True), 
+        desc=f"{datetime.now().strftime('%Y-%m-%d %H:%M:%S.%f')[:-3]} | INFO     | {self.log_label}Extracting isotope patterns"
+    )):
+        feature_rt = row["rt"]
+        feature_mz = row["mz"]
+        
+        # Find MS1 scans within RT tolerance
+        rt_mask = (
+            (self.ms1_df["rt"] >= (feature_rt - rt_tolerance_s)) & 
+            (self.ms1_df["rt"] <= (feature_rt + rt_tolerance_s))
+        )
+        ms1_in_range = self.ms1_df.filter(rt_mask)
+        
+        if ms1_in_range.is_empty():
+            ms1_specs.append(None)
+            feature_indices.append(row["feature_uid"])
+            continue
+        
+        # Extract isotopic pattern
+        isotope_pattern = []
+        
+        # Start with the monoisotopic peak (M+0)
+        base_intensity = 0
+        mz_tolerance = 0.01  # 10 ppm at 1000 Da
+        
+        # Find the base peak intensity
+        base_mask = (
+            (ms1_in_range["mz"] >= (feature_mz - mz_tolerance)) & 
+            (ms1_in_range["mz"] <= (feature_mz + mz_tolerance))
+        )
+        base_peaks = ms1_in_range.filter(base_mask)
+        
+        if not base_peaks.is_empty():
+            base_intensity = base_peaks["inty"].max()
+            isotope_pattern.append([feature_mz, base_intensity])
+        
+        # Look for isotope peaks
+        for shift in isotope_shifts:
+            isotope_mz = feature_mz + shift
+            isotope_mask = (
+                (ms1_in_range["mz"] >= (isotope_mz - mz_tolerance)) & 
+                (ms1_in_range["mz"] <= (isotope_mz + mz_tolerance))
+            )
+            isotope_peaks = ms1_in_range.filter(isotope_mask)
+            
+            if not isotope_peaks.is_empty():
+                max_intensity = isotope_peaks["inty"].max()
+                # Only keep isotope peaks that are at least 1% of base peak
+                if base_intensity > 0 and max_intensity >= 0.01 * base_intensity:
+                    # Get the mz of the most intense peak
+                    max_peak = isotope_peaks.filter(pl.col("inty") == max_intensity).row(0, named=True)
+                    isotope_pattern.append([max_peak["mz"], max_intensity])
+        
+        # Convert to numpy array or None if empty
+        if len(isotope_pattern) > 1:  # Need at least 2 points (monoisotopic + 1 isotope)
+            ms1_spec = np.array(isotope_pattern, dtype=np.float64)
+        else:
+            ms1_spec = None
+            
+        ms1_specs.append(ms1_spec)
+        feature_indices.append(row["feature_uid"])
+    
+    # Update the features_df with the isotopic spectra
+    update_df = pl.DataFrame({
+        "feature_uid": feature_indices,
+        "ms1_spec_new": pl.Series("ms1_spec_new", ms1_specs, dtype=pl.Object)
+    })
+    
+    # Join and update
+    self.features_df = (
+        self.features_df.join(
+            update_df,
+            on="feature_uid",
+            how="left"
+        )
+        .with_columns([
+            pl.when(pl.col("ms1_spec_new").is_not_null())
+            .then(pl.col("ms1_spec_new"))
+            .otherwise(pl.col("ms1_spec"))
+            .alias("ms1_spec")
+        ])
+        .drop("ms1_spec_new")
+    )
+    
+    # Log results
+    non_null_count = len([spec for spec in ms1_specs if spec is not None])
+    self.logger.info(f"Extracted isotopic distributions for {non_null_count}/{len(ms1_specs)} features.")
+    
+    # Store parameters in history
+    params_dict = {"rt_tolerance": rt_tolerance}
+    params_dict.update(kwargs)
+    self.store_history(["find_iso"], params_dict)

masster/sample/sample.py

@@ -97,6 +97,7 @@ from masster.sample.processing import _get_ztscan_stats
 from masster.sample.processing import _spec_to_mat
 from masster.sample.processing import analyze_dda
 from masster.sample.processing import find_features
+from masster.sample.processing import find_iso
 from masster.sample.processing import find_ms2
 from masster.sample.processing import get_spectrum
 from masster.sample.parameters import store_history
@@ -218,6 +219,7 @@ class Sample:
     save = save
     find_features = find_features
     find_adducts = find_adducts
+    find_iso = find_iso
     find_ms2 = find_ms2
     get_spectrum = get_spectrum
     filter = features_filter

masster/sample/sample5_schema.json

@@ -90,6 +90,9 @@
       },
       "ms2_specs": {
         "dtype": "pl.Object"
+      },
+      "ms1_spec": {
+        "dtype": "pl.Object"
       }
     }
   },