masster 0.5.13__tar.gz → 0.5.14__tar.gz

{masster-0.5.13 → masster-0.5.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: masster
-Version: 0.5.13
+Version: 0.5.14
 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-0.5.13 → masster-0.5.14}/pyproject.toml

@@ -1,7 +1,7 @@
 
 [project]
 name = "masster"
-version = "0.5.13"
+version = "0.5.14"
 description = "Mass spectrometry data analysis package"
 authors = [
     { name = "Zamboni Lab" }

{masster-0.5.13 → masster-0.5.14}/src/masster/_version.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 
-__version__ = "0.5.13"
+__version__ = "0.5.14"
 
 
 def get_version():

{masster-0.5.13 → masster-0.5.14}/src/masster/lib/lib.py

@@ -46,11 +46,9 @@ annotations = lib.annotate_features(sample.features_df)
 """
 
 import os
-import uuid
 from typing import Optional, Union, List, Dict, Any, TYPE_CHECKING
 import warnings
 
-import numpy as np
 import polars as pl
 import pyopenms as oms
 
@@ -58,6 +56,148 @@ if TYPE_CHECKING:
     import pandas as pd
 
 
+def _calculate_formula_mass_shift(formula: str) -> float:
+    """
+    Calculate mass shift from formula string like "+H", "-H2O", "+Na-H", etc.
+
+    Parameters
+    ----------
+    formula : str
+        Formula string (e.g., "+H", "-H2O", "+Na-H")
+
+    Returns
+    -------
+    float
+        Mass shift in Daltons
+    """
+    # Standard atomic masses
+    atomic_masses = {
+        "H": 1.007825,
+        "C": 12.0,
+        "N": 14.003074,
+        "O": 15.994915,
+        "Na": 22.989769,
+        "K": 38.963707,
+        "Li": 7.016003,
+        "Ca": 39.962591,
+        "Mg": 23.985042,
+        "Fe": 55.934938,
+        "Cl": 34.968853,
+        "Br": 78.918336,
+        "I": 126.904473,
+        "P": 30.973762,
+        "S": 31.972071,
+    }
+
+    total_mass = 0.0
+
+    # Parse formula by splitting on + and - while preserving the operators
+    parts = []
+    current_part = ""
+    current_sign = 1
+
+    for char in formula:
+        if char == "+":
+            if current_part:
+                parts.append((current_sign, current_part))
+            current_part = ""
+            current_sign = 1
+        elif char == "-":
+            if current_part:
+                parts.append((current_sign, current_part))
+            current_part = ""
+            current_sign = -1
+        else:
+            current_part += char
+
+    if current_part:
+        parts.append((current_sign, current_part))
+
+    # Process each part
+    for sign, part in parts:
+        if not part:
+            continue
+
+        # Parse element and count (e.g., "H2O" -> H:2, O:1)
+        elements = _parse_element_counts(part)
+
+        for element, count in elements.items():
+            if element in atomic_masses:
+                total_mass += sign * atomic_masses[element] * count
+
+    return total_mass
+
+
+def _parse_element_counts(formula_part: str) -> Dict[str, int]:
+    """Parse element counts from a formula part like 'H2O' -> {'H': 2, 'O': 1}"""
+    elements = {}
+    i = 0
+
+    while i < len(formula_part):
+        # Get element (uppercase letter, possibly followed by lowercase)
+        element = formula_part[i]
+        i += 1
+
+        while i < len(formula_part) and formula_part[i].islower():
+            element += formula_part[i]
+            i += 1
+
+        # Get count (digits following element)
+        count_str = ""
+        while i < len(formula_part) and formula_part[i].isdigit():
+            count_str += formula_part[i]
+            i += 1
+
+        count = int(count_str) if count_str else 1
+        elements[element] = elements.get(element, 0) + count
+
+    return elements
+
+
+def _format_adduct_name(components: List[Dict]) -> str:
+    """Format adduct name from components like [M+H]1+ or [M+2H]2+"""
+    if not components:
+        return "[M]"
+
+    # Count occurrences of each formula
+    from collections import Counter
+
+    formula_counts = Counter(comp["formula"] for comp in components)
+    total_charge = sum(comp["charge"] for comp in components)
+
+    # Build formula part with proper multipliers
+    formula_parts = []
+    for formula, count in sorted(
+        formula_counts.items(),
+    ):  # Sort for consistent ordering
+        if count == 1:
+            formula_parts.append(formula)
+        else:
+            # For multiple occurrences, use count prefix (e.g., 2H, 3Na)
+            # Handle special case where formula might already start with + or -
+            if formula.startswith(("+", "-")):
+                sign = formula[0]
+                base_formula = formula[1:]
+                formula_parts.append(f"{sign}{count}{base_formula}")
+            else:
+                formula_parts.append(f"{count}{formula}")
+
+    # Combine formula parts
+    formula = "".join(formula_parts)
+
+    # Format charge
+    if total_charge == 0:
+        charge_str = ""
+    elif abs(total_charge) == 1:
+        charge_str = "1+" if total_charge > 0 else "1-"
+    else:
+        charge_str = (
+            f"{abs(total_charge)}+" if total_charge > 0 else f"{abs(total_charge)}-"
+        )
+
+    return f"[M{formula}]{charge_str}"
+
+
 class Lib:
     """
     A class for managing compound libraries and feature annotation in mass spectrometry data.
@@ -89,22 +229,21 @@ class Lib:
         >>> print(f"Loaded {len(lib.lib_df)} library entries")
     """
     
-    # Define supported adducts and their properties
-    ADDUCT_DEFINITIONS = {
-        # Positive mode adducts
-        "[M+H]1+": {"delta_m": 1.007276, "delta_z": 1, "polarity": "positive"},
-        "[M+Na]1+": {"delta_m": 22.989218, "delta_z": 1, "polarity": "positive"},
-        "[M+K]1+": {"delta_m": 38.962383, "delta_z": 1, "polarity": "positive"},
-        "[M+NH4]1+": {"delta_m": 18.033823, "delta_z": 1, "polarity": "positive"},
-        "[M+H-H2O]1+": {"delta_m": -17.00329, "delta_z": 1, "polarity": "positive"},
-        "[M+2H]2+": {"delta_m": 2.014552, "delta_z": 2, "polarity": "positive"},
-        
-        # Negative mode adducts
-        "[M-H]1-": {"delta_m": -1.007276, "delta_z": -1, "polarity": "negative"},
-        "[M+CH3COO]1-": {"delta_m": 59.013852, "delta_z": -1, "polarity": "negative"},
-        "[M+HCOO]1-": {"delta_m": 44.998203, "delta_z": -1, "polarity": "negative"},
-        "[M+Cl]1-": {"delta_m": 34.968853, "delta_z": -1, "polarity": "negative"},
-        "[M-2H]2-": {"delta_m": -2.014552, "delta_z": -2, "polarity": "negative"},
+    # Default adduct definitions using OpenMS format
+    DEFAULT_ADDUCTS = {
+        "positive": [
+            "+H:1:0.65",
+            "+Na:1:0.15",
+            "+K:1:0.05",
+            "+NH4:1:0.15",
+            "-H2O:0:0.15",
+        ],
+        "negative": [
+            "-H:-1:0.9",
+            "+Cl:-1:0.1",
+            "+CH2O2:0:0.15",
+            "-H2O:0:0.15",
+        ]
     }
     
     def __init__(self):
@@ -119,12 +258,15 @@ class Lib:
             "cmpd_uid": pl.Series([], dtype=pl.Int64),
             "source_id": pl.Series([], dtype=pl.Utf8),
             "name": pl.Series([], dtype=pl.Utf8),
+            "shortname": pl.Series([], dtype=pl.Utf8),
+            "class": pl.Series([], dtype=pl.Utf8),
             "smiles": pl.Series([], dtype=pl.Utf8),
             "inchi": pl.Series([], dtype=pl.Utf8),
             "inchikey": pl.Series([], dtype=pl.Utf8),
             "formula": pl.Series([], dtype=pl.Utf8),
             "iso": pl.Series([], dtype=pl.Int64),
             "adduct": pl.Series([], dtype=pl.Utf8),
+            "probability": pl.Series([], dtype=pl.Float64),
             "m": pl.Series([], dtype=pl.Float64),
             "z": pl.Series([], dtype=pl.Int8),
             "mz": pl.Series([], dtype=pl.Float64),
@@ -134,6 +276,177 @@ class Lib:
             "db": pl.Series([], dtype=pl.Utf8),
         })
     
+    def _get_adducts(self, 
+                    adducts_list: Optional[List[str]] = None, 
+                    polarity: Optional[str] = None,
+                    min_probability: float = 0.03,
+                    **kwargs) -> pl.DataFrame:
+        """
+        Generate comprehensive adduct specifications for the library.
+        
+        This method creates a DataFrame of adduct combinations following the same
+        syntax as Study() and Sample() classes.
+        
+        Args:
+            adducts_list: List of adduct specifications in OpenMS format (e.g., "+H:1:0.65")
+            polarity: "positive", "negative", or None for both
+            min_probability: Minimum probability threshold to filter adducts
+            **kwargs: Additional parameters for adduct generation
+            
+        Returns:
+            DataFrame with columns:
+            - name: Formatted adduct name like "[M+H]1+"
+            - charge: Total charge of the adduct
+            - mass_shift: Total mass shift in Da
+            - probability: Combined probability score
+            - complexity: Number of adduct components
+        """
+        # Get adduct specifications
+        if adducts_list is None:
+            if polarity is None:
+                # Use positive by default
+                adducts_list = self.DEFAULT_ADDUCTS["positive"]
+            elif polarity.lower() in ["positive", "pos"]:
+                adducts_list = self.DEFAULT_ADDUCTS["positive"]
+            elif polarity.lower() in ["negative", "neg"]:
+                adducts_list = self.DEFAULT_ADDUCTS["negative"]
+            else:
+                raise ValueError(f"Unknown polarity: {polarity}")
+        
+        # Parameters
+        charge_min = kwargs.get("charge_min", -2)
+        charge_max = kwargs.get("charge_max", 2)
+        max_combinations = kwargs.get("max_combinations", 2)
+        
+        # Parse base adduct specifications
+        base_specs = []
+        
+        for adduct_str in adducts_list:
+            if not isinstance(adduct_str, str) or ":" not in adduct_str:
+                continue
+                
+            try:
+                parts = adduct_str.split(":")
+                if len(parts) != 3:
+                    continue
+                    
+                formula_part = parts[0]
+                charge = int(parts[1])
+                probability = float(parts[2])
+                
+                # Calculate mass shift from formula
+                mass_shift = _calculate_formula_mass_shift(formula_part)
+                
+                base_specs.append({
+                    "formula": formula_part,
+                    "charge": charge,
+                    "mass_shift": mass_shift,
+                    "probability": probability,
+                    "raw_string": adduct_str,
+                })
+                
+            except (ValueError, IndexError):
+                continue
+        
+        if not base_specs:
+            # Return empty DataFrame with correct schema
+            return pl.DataFrame({
+                "name": [],
+                "charge": [],
+                "mass_shift": [],
+                "probability": [],
+                "complexity": [],
+            })
+        
+        # Generate all valid combinations
+        combinations_list = []
+        
+        # Separate specs by charge type
+        positive_specs = [spec for spec in base_specs if spec["charge"] > 0]
+        negative_specs = [spec for spec in base_specs if spec["charge"] < 0]
+        neutral_specs = [spec for spec in base_specs if spec["charge"] == 0]
+        
+        # 1. Single adducts
+        for spec in base_specs:
+            if charge_min <= spec["charge"] <= charge_max:
+                formatted_name = _format_adduct_name([spec])
+                combinations_list.append({
+                    "components": [spec],
+                    "formatted_name": formatted_name,
+                    "total_mass_shift": spec["mass_shift"],
+                    "total_charge": spec["charge"],
+                    "combined_probability": spec["probability"],
+                    "complexity": 1,
+                })
+        
+        # 2. Generate multiply charged versions (2H+, 3H+, etc.)
+        if max_combinations >= 2:
+            for spec in positive_specs + negative_specs:
+                base_charge = spec["charge"]
+                for multiplier in range(2, min(max_combinations + 1, 4)):
+                    total_charge = base_charge * multiplier
+                    if charge_min <= total_charge <= charge_max:
+                        components = [spec] * multiplier
+                        formatted_name = _format_adduct_name(components)
+                        
+                        combinations_list.append({
+                            "components": components,
+                            "formatted_name": formatted_name,
+                            "total_mass_shift": spec["mass_shift"] * multiplier,
+                            "total_charge": total_charge,
+                            "combined_probability": spec["probability"] ** multiplier,
+                            "complexity": multiplier,
+                        })
+        
+        # 3. Mixed combinations (positive + neutral)
+        if max_combinations >= 2:
+            for pos_spec in positive_specs:
+                for neut_spec in neutral_specs:
+                    total_charge = pos_spec["charge"] + neut_spec["charge"]
+                    if charge_min <= total_charge <= charge_max:
+                        components = [pos_spec, neut_spec]
+                        formatted_name = _format_adduct_name(components)
+                        combinations_list.append({
+                            "components": components,
+                            "formatted_name": formatted_name,
+                            "total_mass_shift": pos_spec["mass_shift"] + neut_spec["mass_shift"],
+                            "total_charge": total_charge,
+                            "combined_probability": pos_spec["probability"] * neut_spec["probability"],
+                            "complexity": 2,
+                        })
+        
+        # Convert to polars DataFrame
+        if combinations_list:
+            combinations_list.sort(
+                key=lambda x: (-x["combined_probability"], x["complexity"])
+            )
+            
+            adducts_df = pl.DataFrame([
+                {
+                    "name": combo["formatted_name"],
+                    "charge": combo["total_charge"],
+                    "mass_shift": combo["total_mass_shift"],
+                    "probability": combo["combined_probability"],
+                    "complexity": combo["complexity"],
+                }
+                for combo in combinations_list
+            ])
+        else:
+            # Return empty DataFrame with correct schema
+            adducts_df = pl.DataFrame({
+                "name": [],
+                "charge": [],
+                "mass_shift": [],
+                "probability": [],
+                "complexity": [],
+            })
+        
+        # Filter by minimum probability
+        if min_probability > 0.0 and len(adducts_df) > 0:
+            adducts_df = adducts_df.filter(pl.col("probability") >= min_probability)
+        
+        return adducts_df
+    
     def _calculate_accurate_mass(self, formula: str) -> Optional[float]:
         """
         Calculate the accurate mass for a molecular formula using PyOpenMS.
@@ -185,15 +498,17 @@ class Lib:
                                 compound_data: Dict[str, Any], 
                                 adducts: Optional[List[str]] = None,
                                 polarity: Optional[str] = None,
-                                lib_id_counter: Optional[int] = None) -> tuple[List[Dict[str, Any]], int]:
+                                lib_id_counter: Optional[int] = None,
+                                min_probability: float = 0.03) -> tuple[List[Dict[str, Any]], int]:
         """
-        Generate adduct variants for a given compound.
+        Generate adduct variants for a given compound using the new adduct system.
         
         Args:
             compound_data: Dictionary containing compound information
-            adducts: List of specific adducts to generate. If None, uses all adducts for polarity
-            polarity: Ionization polarity ("positive", "negative", or None for both)
+            adducts: List of specific adducts to generate. If None, uses defaults for polarity
+            polarity: Ionization polarity ("positive", "negative", or None for positive)
             lib_id_counter: Counter for generating unique lib_uid values
+            min_probability: Minimum probability threshold for adduct filtering
             
         Returns:
             Tuple of (list of dictionaries representing adduct variants, updated counter)
@@ -206,35 +521,25 @@ class Lib:
         if accurate_mass is None:
             return variants, counter
         
-        # Determine which adducts to use
-        if adducts is None:
-            if polarity is None:
-                # Use all adducts
-                selected_adducts = list(self.ADDUCT_DEFINITIONS.keys())
-            else:
-                # Filter by polarity
-                selected_adducts = [
-                    adduct for adduct, props in self.ADDUCT_DEFINITIONS.items()
-                    if props["polarity"] == polarity.lower()
-                ]
-        else:
-            selected_adducts = adducts
+        # Get adduct specifications using _get_adducts
+        adducts_df = self._get_adducts(
+            adducts_list=adducts,
+            polarity=polarity,
+            min_probability=min_probability
+        )
+        
+        if len(adducts_df) == 0:
+            return variants, counter
         
         # Generate variants for each adduct
-        for adduct in selected_adducts:
-            if adduct not in self.ADDUCT_DEFINITIONS:
-                warnings.warn(f"Unknown adduct: {adduct}")
-                continue
-                
-            adduct_props = self.ADDUCT_DEFINITIONS[adduct]
-            
-            # Skip if polarity doesn't match
-            if polarity is not None and adduct_props["polarity"] != polarity.lower():
-                continue
+        for adduct_row in adducts_df.iter_rows(named=True):
+            adduct_name = adduct_row["name"]
+            charge = adduct_row["charge"]
+            mass_shift = adduct_row["mass_shift"]
+            probability = adduct_row["probability"]
             
             # Calculate adducted mass and m/z
-            adducted_mass = accurate_mass + adduct_props["delta_m"]
-            charge = adduct_props["delta_z"]
+            adducted_mass = accurate_mass + mass_shift
             mz = abs(adducted_mass / charge) if charge != 0 else adducted_mass
             
             # Create variant entry
@@ -243,12 +548,15 @@ class Lib:
                 "cmpd_uid": compound_data.get("cmpd_uid", None),
                 "source_id": compound_data.get("source_id", None),
                 "name": compound_data.get("name", ""),
+                "shortname": compound_data.get("shortname", ""),
+                "class": compound_data.get("class", ""),
                 "smiles": compound_data.get("smiles", ""),
                 "inchi": compound_data.get("inchi", ""),
                 "inchikey": compound_data.get("inchikey", ""),
                 "formula": compound_data["formula"],
                 "iso": 0,  # Default to zero
-                "adduct": adduct,
+                "adduct": adduct_name,
+                "probability": probability,
                 "m": adducted_mass,
                 "z": charge,
                 "mz": mz,
@@ -265,7 +573,8 @@ class Lib:
     def import_csv(self, 
                   csvfile: str, 
                   polarity: Optional[str] = None,
-                  adducts: Optional[List[str]] = None) -> None:
+                  adducts: Optional[List[str]] = None,
+                  min_probability: float = 0.03) -> None:
         """
         Import compound library from a CSV file.
         
@@ -274,8 +583,9 @@ class Lib:
         
         Args:
             csvfile: Path to the CSV file
-            polarity: Ionization polarity ("positive", "negative", or None for both)
-            adducts: Specific adducts to generate. If None, generates all for the polarity
+            polarity: Ionization polarity ("positive", "negative", or None for positive)
+            adducts: Specific adducts to generate. If None, generates defaults for the polarity
+            min_probability: Minimum probability threshold for adduct filtering
             
         Expected CSV columns (case-insensitive):
             - Required: Formula (or formula)
@@ -319,6 +629,8 @@ class Lib:
 
             compound_data = {
                 "name": row.get(column_mapping.get("name", ""), ""),
+                "shortname": row.get(column_mapping.get("shortname", ""), ""),
+                "class": row.get(column_mapping.get("class", ""), ""),                
                 "smiles": row.get(column_mapping.get("smiles", ""), ""),
                 "inchi": row.get(column_mapping.get("inchi", ""), ""),
                 "inchikey": row.get(column_mapping.get("inchikey", ""), ""),
@@ -331,7 +643,8 @@ class Lib:
             
             # Generate adduct variants
             variants, lib_id_counter = self._generate_adduct_variants(
-                compound_data, adducts=adducts, polarity=polarity, lib_id_counter=lib_id_counter
+                compound_data, adducts=adducts, polarity=polarity, 
+                lib_id_counter=lib_id_counter, min_probability=min_probability
             )
             all_variants.extend(variants)
             
@@ -349,7 +662,8 @@ class Lib:
                     compound_data_rt2["name"] = compound_data["name"] + " II"
                     
                     variants_rt2, lib_id_counter = self._generate_adduct_variants(
-                        compound_data_rt2, adducts=adducts, polarity=polarity, lib_id_counter=lib_id_counter
+                        compound_data_rt2, adducts=adducts, polarity=polarity, 
+                        lib_id_counter=lib_id_counter, min_probability=min_probability
                     )
                     all_variants.extend(variants_rt2)
         
@@ -529,6 +843,8 @@ class Lib:
                     "cmpd_uid": match_row.get("cmpd_uid"),
                     "source_id": match_row.get("source_id"),
                     "name": match_row["name"],
+                    "shortname": match_row["shortname"],
+                    "class": match_row["class"],
                     "formula": match_row["formula"],
                     "iso": match_row.get("iso", 0),
                     "adduct": match_row["adduct"],
@@ -555,10 +871,8 @@ class Lib:
         Returns:
             List of adduct names
         """
-        return [
-            adduct for adduct, props in self.ADDUCT_DEFINITIONS.items()
-            if props["polarity"] == polarity.lower()
-        ]
+        adducts_df = self._get_adducts(polarity=polarity, min_probability=0.0)
+        return adducts_df.select("name").to_series().to_list()
     
     def __len__(self) -> int:
         """Return number of library entries."""

{masster-0.5.13 → masster-0.5.14}/src/masster/study/helpers.py

@@ -490,6 +490,7 @@ def align_reset(self):
     # Ensure column order is maintained after with_columns operation
     from masster.study.helpers import _ensure_features_df_schema_order
     _ensure_features_df_schema_order(self)
+    self.logger.info("Alignment reset: all feature RTs set to original_RT.")
 
 
 # =====================================================================================