masster 0.3.17__tar.gz → 0.3.18__tar.gz

{masster-0.3.17 → masster-0.3.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: masster
-Version: 0.3.17
+Version: 0.3.18
 Summary: Mass spectrometry data analysis package
 Project-URL: homepage, https://github.com/zamboni-lab/masster
 Project-URL: repository, https://github.com/zamboni-lab/masster

{masster-0.3.17 → masster-0.3.18}/pyproject.toml

@@ -1,7 +1,7 @@
 
 [project]
 name = "masster"
-version = "0.3.17"
+version = "0.3.18"
 description = "Mass spectrometry data analysis package"
 authors = [
     { name = "Zamboni Lab" }

{masster-0.3.17 → masster-0.3.18}/src/masster/_version.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 
-__version__ = "0.3.17"
+__version__ = "0.3.18"
 
 
 def get_version():

{masster-0.3.17 → masster-0.3.18}/src/masster/study/helpers.py

@@ -479,7 +479,9 @@ def get_consensus(self, quant="chrom_area"):
     # sort by consensus_id
     df1 = df1.sort_index()
 
-    df2 = self.get_consensus_matrix(quant=quant)
+    df2_polars = self.get_consensus_matrix(quant=quant)
+    # Convert to pandas for merging (since the result is used for export)
+    df2 = df2_polars.to_pandas().set_index("consensus_uid")
     # sort df2 row by consensus_id
     df2 = df2.sort_index()
     # merge df and df2 on consensus_id
@@ -492,6 +494,7 @@ def get_consensus(self, quant="chrom_area"):
 def get_consensus_matrix(self, quant="chrom_area"):
     """
     Get a matrix of consensus features with samples as columns and consensus features as rows.
+    Optimized implementation that avoids expensive join operations.
     """
     if quant not in self.features_df.columns:
         self.logger.error(
@@ -499,41 +502,58 @@ def get_consensus_matrix(self, quant="chrom_area"):
         )
         return None
 
-    # Use Polars join instead of pandas merge
-    features_subset = self.features_df.select(["feature_uid", "sample_uid", quant])
-    consensus_mapping_subset = self.consensus_mapping_df.select([
-        "consensus_uid",
-        "feature_uid",
-    ])
-
-    df1 = features_subset.join(
-        consensus_mapping_subset,
-        on="feature_uid",
-        how="left",
-    )
-
-    # Convert to pandas for pivot operation (Polars pivot is still evolving)
-    df1_pd = df1.to_pandas()
-    df2 = df1_pd.pivot_table(
-        index="consensus_uid",
-        columns="sample_uid",
-        values=quant,
-        aggfunc="max",
-    )
-
-    # Create sample_uid to sample_name mapping using Polars
-    sample_mapping = dict(
-        self.samples_df.select(["sample_uid", "sample_name"]).iter_rows(),
-    )
-    # replace sample_uid with sample_name in df2
-    df2 = df2.rename(columns=sample_mapping)
+    # Create a lookup dictionary from features_df for O(1) value access
+    feature_values = {}
+    for row in self.features_df.iter_rows(named=True):
+        feature_uid = row['feature_uid']
+        sample_uid = row['sample_uid']
+        value = row[quant] if row[quant] is not None else 0
+        feature_values[(feature_uid, sample_uid)] = value
+
+    # Build consensus matrix directly using the consensus_mapping_df
+    matrix_dict = {}
+    sample_mapping = dict(self.samples_df.select(["sample_uid", "sample_name"]).iter_rows())
+    
+    for row in self.consensus_mapping_df.iter_rows(named=True):
+        consensus_uid = row['consensus_uid']
+        sample_uid = row['sample_uid']
+        feature_uid = row['feature_uid']
+        
+        # Look up the quantification value
+        key = (feature_uid, sample_uid)
+        value = feature_values.get(key, 0)
+            
+        if consensus_uid not in matrix_dict:
+            matrix_dict[consensus_uid] = {}
+            
+        sample_name = sample_mapping.get(sample_uid, f"sample_{sample_uid}")
+        
+        # Take max if multiple features map to same consensus/sample combination
+        if sample_name in matrix_dict[consensus_uid]:
+            matrix_dict[consensus_uid][sample_name] = max(matrix_dict[consensus_uid][sample_name], value)
+        else:
+            matrix_dict[consensus_uid][sample_name] = value
 
-    # round to integer
-    df2 = df2.round()
-    # set consensus_id as uint64
-    df2.index = df2.index.astype("uint64")
-    # set index to consensus_id
-    df2.index.name = "consensus_uid"
+    # Convert to Polars DataFrame with proper formatting
+    import polars as pl
+    
+    # Convert matrix_dict to list of records for Polars
+    records = []
+    for consensus_uid, sample_values in matrix_dict.items():
+        record = {"consensus_uid": consensus_uid}
+        record.update(sample_values)
+        records.append(record)
+    
+    # Create Polars DataFrame and set proper data types
+    df2 = pl.DataFrame(records)
+    
+    # Fill null values with 0 and round numeric columns
+    numeric_cols = [col for col in df2.columns if col != "consensus_uid"]
+    df2 = df2.with_columns([
+        pl.col("consensus_uid").cast(pl.UInt64),
+        *[pl.col(col).fill_null(0).round(0) for col in numeric_cols]
+    ])
+    
     return df2
 
 

{masster-0.3.17 → masster-0.3.18}/src/masster/study/load.py

@@ -1379,7 +1379,7 @@ def _add_sample_optimized(self, file, type=None, reset=False, adducts=None, skip
     self.samples_df = pl.concat([self.samples_df, new_sample])
     
     # SIMPLIFIED feature processing
-    current_sample_uid = len(self.samples_df) - 1
+    current_sample_uid = len(self.samples_df)
     
     # Add required columns with minimal operations
     columns_to_add = [
@@ -1520,7 +1520,7 @@ def _add_sample_standard(self, file, type=None, reset=False, adducts=None, skip_
     self.samples_df = pl.concat([self.samples_df, new_sample])
     
     # SIMPLIFIED feature processing
-    current_sample_uid = len(self.samples_df) - 1
+    current_sample_uid = len(self.samples_df)
     
     # Add required columns with minimal operations
     columns_to_add = [
@@ -1621,7 +1621,7 @@ def _add_sample_standard(self, file, type=None, reset=False, adducts=None, skip_
     self.samples_df = pl.concat([self.samples_df, new_sample])
     
     # SIMPLIFIED feature processing
-    current_sample_uid = len(self.samples_df) - 1
+    current_sample_uid = len(self.samples_df)
     
     # Add required columns with minimal operations
     columns_to_add = [
@@ -1695,3 +1695,4 @@ def _sample_color_reset_optimized(self):
     )
     
     self.logger.debug(f"Reset sample colors (cached) for {n_samples} samples")
+

{masster-0.3.17 → masster-0.3.18}/src/masster/study/plot.py

@@ -17,7 +17,7 @@ hv.extension("bokeh")
 from bokeh.layouts import row as bokeh_row
 
 
-def plot_alignment(self, maps: bool = True, filename: str | None = None, width: int = 450, height: int = 450, markersize: int = 3):
+def plot_alignment(self, maps: bool = True, samples: int | list[int | str] | None = None, filename: str | None = None, width: int = 450, height: int = 450, markersize: int = 3):
     """Visualize retention time alignment using two synchronized Bokeh scatter plots.
 
     - When ``maps=True`` the function reads ``self.features_maps`` (list of FeatureMap)
@@ -27,6 +27,11 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
 
     Parameters
     - maps: whether to use feature maps (default True).
+    - samples: Sample selection parameter, interpreted like in plot_samples_2d:
+        - None: show all samples
+        - int: show a random subset of N samples  
+        - list of ints: show samples with these sample_uids
+        - list of strings: show samples with these sample_names
     - filename: optional HTML file path to save the plot.
     - width/height: pixel size of each subplot.
     - markersize: base marker size.
@@ -54,6 +59,32 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
             self.logger.error("No feature maps available for plotting.")
             return
 
+        # Get sample_uids to limit which samples to show
+        sample_uids_to_show = self._get_sample_uids(samples)
+        
+        # Filter feature maps based on sample selection
+        if sample_uids_to_show is not None:
+            # Get sample indices for the selected sample_uids
+            selected_indices = []
+            if hasattr(self, 'samples_df') and self.samples_df is not None and not self.samples_df.is_empty():
+                samples_info = self.samples_df.to_pandas()
+                for idx, row in samples_info.iterrows():
+                    if row.get('sample_uid') in sample_uids_to_show:
+                        selected_indices.append(idx)
+            else:
+                # If no samples_df, just limit to the first N samples
+                if isinstance(samples, int):
+                    selected_indices = list(range(min(samples, len(fmaps))))
+                else:
+                    selected_indices = list(range(len(fmaps)))
+            
+            # Filter feature maps to only include selected indices
+            fmaps = [fmaps[i] for i in selected_indices if i < len(fmaps)]
+            
+            if not fmaps:
+                self.logger.error("No feature maps match the selected samples.")
+                return
+
         # Reference (first) sample: use current RT for both before and after
         ref = fmaps[0]
         ref_rt = [f.getRT() for f in ref]
@@ -143,6 +174,28 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
             self.logger.error("No sample identifier column found in features_df.")
             return
 
+        # Get sample_uids to limit which samples to show
+        sample_uids_to_show = self._get_sample_uids(samples)
+        
+        # Filter features_df based on sample selection if specified
+        if sample_uids_to_show is not None:
+            if sample_col == 'sample_uid':
+                features_df = features_df.filter(pl.col('sample_uid').is_in(sample_uids_to_show))
+            else:
+                # Need to convert sample names to sample_uids if using sample_name column
+                if 'sample_uid' in features_df.columns:
+                    # Filter by sample_uid even though we're using sample_name as the primary column
+                    features_df = features_df.filter(pl.col('sample_uid').is_in(sample_uids_to_show))
+                else:
+                    # Convert sample_uids to sample_names and filter
+                    sample_names_to_show = []
+                    if hasattr(self, 'samples_df') and self.samples_df is not None:
+                        for uid in sample_uids_to_show:
+                            matching_rows = self.samples_df.filter(pl.col("sample_uid") == uid)
+                            if not matching_rows.is_empty():
+                                sample_names_to_show.append(matching_rows.row(0, named=True)["sample_name"])
+                    features_df = features_df.filter(pl.col('sample_name').is_in(sample_names_to_show))
+
         # Get unique samples using Polars
         samples = features_df.select(pl.col(sample_col)).unique().to_series().to_list()
 
@@ -1649,11 +1702,19 @@ def plot_pca(
 
     self.logger.debug(f"Performing PCA on consensus matrix with shape: {consensus_matrix.shape}")
 
-    # Convert consensus matrix to numpy if it's not already
-    if hasattr(consensus_matrix, "values"):
+    # Convert consensus matrix to numpy - handle both Polars and pandas DataFrames
+    if hasattr(consensus_matrix, "to_numpy"):
+        # Polars or pandas DataFrame
+        if hasattr(consensus_matrix, "select"):
+            # Polars DataFrame - exclude the consensus_uid column
+            numeric_cols = [col for col in consensus_matrix.columns if col != "consensus_uid"]
+            matrix_data = consensus_matrix.select(numeric_cols).to_numpy()
+        else:
+            # Pandas DataFrame
+            matrix_data = consensus_matrix.to_numpy()
+    elif hasattr(consensus_matrix, "values"):
+        # Pandas DataFrame
         matrix_data = consensus_matrix.values
-    elif hasattr(consensus_matrix, "to_numpy"):
-        matrix_data = consensus_matrix.to_numpy()
     else:
         matrix_data = np.array(consensus_matrix)
 

{masster-0.3.17 → masster-0.3.18}/uv.lock

@@ -1372,7 +1372,7 @@ wheels = [
 
 [[package]]
 name = "masster"
-version = "0.3.17"
+version = "0.3.18"
 source = { editable = "." }
 dependencies = [
     { name = "alphabase" },