corems.molecular_id.factory.spectrum_search_results

  1class SpectrumSearchResults:
  2    """Class for storing Search Results for a single Spectrum Query
  3
  4    Parameters
  5    ----------
  6    query_spectrum : MassSpectrum
  7        The queried mass spectrum
  8    precursor_mz : float, optional
  9        The queried precursor_mz. None is interpreted as an open query, i.e. no precursor_mz
 10    spectral_similarity_search_results : dict
 11        The search results for the queried spectrum, which will be unpacked into attributes
 12
 13    Attributes
 14    ----------
 15    query_spectrum : MassSpectrum
 16        The queried mass spectrum
 17    query_spectrum_id : int
 18        The id of the queried spectrum (the scan number within an MassSpectra object)
 19    precursor_mz : float
 20        The precursor m/z of the queried spectrum
 21
 22    Other Possible Attributes
 23    -------------------------
 24    ref_mol_id : str
 25        The id of the molecule associated with the query spectrum in reference database
 26    ref_ms_id : str
 27        The id of the query spectrum in reference database
 28    ref_precursor_mz : float
 29        The precursor mass of the query spectrum
 30    precursor_mz_error_ppm : float
 31        The ppm error between the query spectrum and the reference spectrum
 32    entropy_similarity : float
 33        The entropy similarity between the query spectrum and the reference spectrum
 34    ref_ion_type : str
 35        The ion type of the reference spectrum, i.e. [M+H]+, [M+Na]+, etc.
 36    query_mz_in_ref_n : list
 37        The number of query m/z peaks that are in the reference spectrum
 38    query_mz_in_ref_fract : float
 39        The fraction of query m/z peaks that are in the reference spectrum
 40    query_frag_types : list
 41        The fragment types of the query spectrum that are in the reference spectrum,
 42        i.e. LSF (lipid species fragments) or MSF (molecular species fragments),
 43        generally used for only for lipidomics
 44    ref_mz_in_query_n : list
 45        The number of reference m/z peaks that are in the query spectrum
 46    ref_mz_in_query_fract : float
 47        The fraction of reference m/z peaks that are in the query spectrum
 48    ref_frag_types : list
 49        The fragment types of the reference spectrum,
 50        i.e. LSF (lipid species fragments) or MSF (molecular species fragments),
 51        generally used for only for lipidomics
 52    database_name : str
 53        The name of the associated database, i.e. "LipidBlast" or "GNPS"
 54
 55    Methods
 56    -------
 57    *to_dataframe().
 58        Convert the SpectrumSearchResults to a pandas DataFrame
 59
 60    """
 61
 62    def __init__(
 63        self, query_spectrum, precursor_mz, spectral_similarity_search_results
 64    ):
 65        self.query_spectrum = query_spectrum
 66        self.precursor_mz = precursor_mz
 67        if query_spectrum is not None:
 68            if query_spectrum.scan_number is not None:
 69                self.query_spectrum_id = query_spectrum.scan_number
 70        attribute_keys = [
 71            "ref_mol_id",
 72            "ref_ms_id",
 73            "ref_precursor_mz",
 74            "precursor_mz_error_ppm",
 75            "ref_ion_type",
 76            "entropy_similarity",
 77            "query_mz_in_ref_n",
 78            "query_mz_in_ref_fract",
 79            "query_frag_types",
 80            "ref_mz_in_query_n",
 81            "ref_mz_in_query_fract",
 82            "ref_frag_types",
 83            "database_name",
 84        ]
 85        for key in spectral_similarity_search_results.keys():
 86            if key in attribute_keys:
 87                setattr(self, key, spectral_similarity_search_results[key])
 88
 89    def to_dataframe(self, cols_to_drop=None):
 90        """Convert the SpectrumSearchResults to a pandas DataFrame
 91
 92        Parameters
 93        ----------
 94        cols_to_drop : list, optional
 95            A list of columns to drop from the DataFrame. Default is None.
 96
 97        Returns
 98        -------
 99        pandas.DataFrame
100            A DataFrame with the SpectrumSearchResults attributes as columns
101
102        """
103        import pandas as pd
104
105        # get initial dict
106        df = pd.DataFrame.from_dict(self.__dict__).copy()
107
108        # remove ms2_spectrum column
109        df = df.drop(columns=["query_spectrum"])
110
111        # drop additional columns
112        if cols_to_drop is not None:
113            df = df.drop(columns=cols_to_drop)
114
115        # reorder to high to low entropy similarity
116        df = df.sort_values(by=["entropy_similarity"], ascending=False)
117
118        # rename id to query_scan_number
119        return pd.DataFrame(df)