msbuddy 0.2.0__tar.gz → 0.2.2__tar.gz

{msbuddy-0.2.0 → msbuddy-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: msbuddy
-Version: 0.2.0
+Version: 0.2.2
 Summary: Bottom-up MS/MS interrogation-based molecular formula annotation for mass spectrometry data.
 Home-page: https://github.com/Philipbear/msbuddy
 Author: Shipei Xing
@@ -20,7 +20,13 @@ License-File: LICENSE
 [![docs](https://readthedocs.org/projects/msbuddy/badge/?version=latest)](https://msbuddy.readthedocs.io/en/latest/?badge=latest)
 [![Generic badge](https://img.shields.io/badge/msbuddy-mass_spec_tools-<COLOR>.svg)](https://github.com/Philipbear/msbuddy)
 
-`msbuddy` is developed for molecular formula annotation in mass spectrometry-based small molecule analysis.
+<p align="center">
+  <img src="https://github.com/Philipbear/msbuddy/blob/main/logo/logo.svg" alt="Sample Image" height="100"/>
+</p>
+
+
+`msbuddy` is developed for molecular formula analysis in MS-based metabolomics studies.
+
 `msbuddy` is able to provide formula annotations for queries:
   * with or without MS1 isotopic pattern 
   * with or without MS/MS spectra
@@ -30,7 +36,7 @@ The minimum requirement for a msbuddy query is a single _m/z_ value and its ioni
 
 **Official documentation**  is available at https://msbuddy.readthedocs.io/en/latest/.
 
-**msbuddy ver 0.2.0** is now released, with refined models and much improved performance. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
+**msbuddy ver 0.2.1** is now released. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
 
 ##  Citation
 > S. Xing et al. BUDDY: molecular formula discovery via bottom-up MS/MS interrogation. **Nature Methods** 2023. [DOI: 10.1038/s41592-023-01850-x](https://doi.org/10.1038/s41592-023-01850-x)
@@ -45,20 +51,21 @@ pip install msbuddy
 
 To install a specific version, see PyPI [release history](https://pypi.org/project/msbuddy/#history).
 
-## Python usage ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
+## Python usage 
 
+### Quick start ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
 As a quick start, we here load a mgf file and annotate the molecular formula for each query.
 All the configurations are specified in the [`MsbuddyConfig`](https://msbuddy.readthedocs.io/en/latest/pyapi.html#msbuddy.MsbuddyConfig) class.
-**Parallel computing** is supported.
+**Parallel computing** is [supported](https://msbuddy.readthedocs.io/en/latest/quickstart.html).
 
 ```python
 from msbuddy import Msbuddy, MsbuddyConfig
 
 # instantiate a MsbuddyConfig object
-msb_config = MsbuddyConfig(ms_instr='orbitrap', # supported: "qtof", "orbitrap" and "fticr"
-                                                # highly recommended to specify
-                           halogen=False, # whether to consider halogen atoms FClBrI
-                           parallel=True, n_cpu=12)
+msb_config = MsbuddyConfig(# highly recommended to specify
+                           ms_instr='orbitrap',  # supported: "qtof", "orbitrap" and "fticr"
+                           # whether to consider halogen atoms FClBrI
+                           halogen=False)
 
 # instantiate a Msbuddy object
 msb_engine = Msbuddy(msb_config)
@@ -72,6 +79,8 @@ msb_engine.annotate_formula()
 # retrieve the annotation result summary
 result = msb_engine.get_summary()
 ```
+See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
+
 
 MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.1101/2020.05.09.086066v2) if they are publicly available:
 ```python
@@ -79,7 +88,14 @@ MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.
 msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
-**msbuddy** also provides python APIs for generating molecular formulas from [a neutral mass](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mass_to_formula) or [a charged _m/z_ value](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mz_to_formula).
+
+### Other API functions
+- generate molecular formulas from a neutral mass (mass_to_formula)
+- generate molecular formulas from a charged _m/z_ value (mz_to_formula)
+- predict formula feasibility using a deep learning model (predict_formula_feasibility)
+- enumerate all possible subformulas of a given precursor formula (enumerate_subform_arr)
+
+See more in the [documentation](https://msbuddy.readthedocs.io/en/latest/pyapi.html).
 
 ## Command-line tool ([docs](https://msbuddy.readthedocs.io/en/latest/cmdapi.html))
 
@@ -95,7 +111,7 @@ for a single USI string (from FT-ICR, for example):
 ```bash
 msbuddy -usi <usi_string> -output <output_folder> -ms fticr
 ```
-or a csv file ([templates files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
+or a csv file ([template files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
 ```bash
 msbuddy -csv <csv_file> -output <output_folder> -ms qtof -details
 ```

{msbuddy-0.2.0 → msbuddy-0.2.2}/README.md

@@ -4,7 +4,13 @@
 [![docs](https://readthedocs.org/projects/msbuddy/badge/?version=latest)](https://msbuddy.readthedocs.io/en/latest/?badge=latest)
 [![Generic badge](https://img.shields.io/badge/msbuddy-mass_spec_tools-<COLOR>.svg)](https://github.com/Philipbear/msbuddy)
 
-`msbuddy` is developed for molecular formula annotation in mass spectrometry-based small molecule analysis.
+<p align="center">
+  <img src="https://github.com/Philipbear/msbuddy/blob/main/logo/logo.svg" alt="Sample Image" height="100"/>
+</p>
+
+
+`msbuddy` is developed for molecular formula analysis in MS-based metabolomics studies.
+
 `msbuddy` is able to provide formula annotations for queries:
   * with or without MS1 isotopic pattern 
   * with or without MS/MS spectra
@@ -14,7 +20,7 @@ The minimum requirement for a msbuddy query is a single _m/z_ value and its ioni
 
 **Official documentation**  is available at https://msbuddy.readthedocs.io/en/latest/.
 
-**msbuddy ver 0.2.0** is now released, with refined models and much improved performance. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
+**msbuddy ver 0.2.1** is now released. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
 
 ##  Citation
 > S. Xing et al. BUDDY: molecular formula discovery via bottom-up MS/MS interrogation. **Nature Methods** 2023. [DOI: 10.1038/s41592-023-01850-x](https://doi.org/10.1038/s41592-023-01850-x)
@@ -29,20 +35,21 @@ pip install msbuddy
 
 To install a specific version, see PyPI [release history](https://pypi.org/project/msbuddy/#history).
 
-## Python usage ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
+## Python usage 
 
+### Quick start ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
 As a quick start, we here load a mgf file and annotate the molecular formula for each query.
 All the configurations are specified in the [`MsbuddyConfig`](https://msbuddy.readthedocs.io/en/latest/pyapi.html#msbuddy.MsbuddyConfig) class.
-**Parallel computing** is supported.
+**Parallel computing** is [supported](https://msbuddy.readthedocs.io/en/latest/quickstart.html).
 
 ```python
 from msbuddy import Msbuddy, MsbuddyConfig
 
 # instantiate a MsbuddyConfig object
-msb_config = MsbuddyConfig(ms_instr='orbitrap', # supported: "qtof", "orbitrap" and "fticr"
-                                                # highly recommended to specify
-                           halogen=False, # whether to consider halogen atoms FClBrI
-                           parallel=True, n_cpu=12)
+msb_config = MsbuddyConfig(# highly recommended to specify
+                           ms_instr='orbitrap',  # supported: "qtof", "orbitrap" and "fticr"
+                           # whether to consider halogen atoms FClBrI
+                           halogen=False)
 
 # instantiate a Msbuddy object
 msb_engine = Msbuddy(msb_config)
@@ -56,6 +63,8 @@ msb_engine.annotate_formula()
 # retrieve the annotation result summary
 result = msb_engine.get_summary()
 ```
+See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
+
 
 MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.1101/2020.05.09.086066v2) if they are publicly available:
 ```python
@@ -63,7 +72,14 @@ MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.
 msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
-**msbuddy** also provides python APIs for generating molecular formulas from [a neutral mass](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mass_to_formula) or [a charged _m/z_ value](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mz_to_formula).
+
+### Other API functions
+- generate molecular formulas from a neutral mass (mass_to_formula)
+- generate molecular formulas from a charged _m/z_ value (mz_to_formula)
+- predict formula feasibility using a deep learning model (predict_formula_feasibility)
+- enumerate all possible subformulas of a given precursor formula (enumerate_subform_arr)
+
+See more in the [documentation](https://msbuddy.readthedocs.io/en/latest/pyapi.html).
 
 ## Command-line tool ([docs](https://msbuddy.readthedocs.io/en/latest/cmdapi.html))
 
@@ -79,7 +95,7 @@ for a single USI string (from FT-ICR, for example):
 ```bash
 msbuddy -usi <usi_string> -output <output_folder> -ms fticr
 ```
-or a csv file ([templates files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
+or a csv file ([template files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
 ```bash
 msbuddy -csv <csv_file> -output <output_folder> -ms qtof -details
 ```

msbuddy-0.2.2/msbuddy/__init__.py

@@ -0,0 +1,3 @@
+from .main import Msbuddy, MsbuddyConfig
+from .utils import read_formula, form_arr_to_str, enumerate_subform_arr
+from .cand import assign_subformula

{msbuddy-0.2.0 → msbuddy-0.2.2}/msbuddy/cand.py

@@ -19,9 +19,9 @@ import numpy as np
 from brainpy import isotopic_variants
 from numba import njit
 
-from msbuddy.base import Formula, CandidateFormula, MS2Explanation, MetaFeature
+from msbuddy.base import Formula, CandidateFormula, MS2Explanation, MetaFeature, check_adduct, Adduct
 from msbuddy.query import check_common_frag, check_common_nl, query_precursor_mass, query_fragnl_mass
-from msbuddy.utils import form_arr_to_str, enumerate_subformula
+from msbuddy.utils import form_arr_to_str, enumerate_subformula, read_formula, SubformulaResult, FormulaResult
 
 
 class FragExplanation:
@@ -705,7 +705,7 @@ def _calc_subform_mass(subform_arr: np.array, adduct_charge: int) -> np.array:
 
 
 @njit
-def _dbe_subform_filter(subform_arr: np.array) -> np.array:
+def _dbe_subform_filter(subform_arr: np.array, cutoff: float) -> np.array:
     """
     Filter subformulas by DBE.
     :param subform_arr: 2D array, each row is a subformula array
@@ -714,7 +714,7 @@ def _dbe_subform_filter(subform_arr: np.array) -> np.array:
     dbe_arr = subform_arr[:, 0] + 1 - (subform_arr[:, 1] + subform_arr[:, 4] + subform_arr[:, 3] + subform_arr[:, 2]
                                        + subform_arr[:, 5] + subform_arr[:, 8] + subform_arr[:, 6]) / 2 + \
               (subform_arr[:, 7] + subform_arr[:, 10]) / 2
-    dbe_bool_arr = dbe_arr >= -1
+    dbe_bool_arr = dbe_arr >= cutoff
     return dbe_bool_arr
 
 
@@ -774,6 +774,7 @@ def _assign_ms2_explanation(mf: MetaFeature, cf: CandidateFormula, pre_charged_a
     :return: CandidateFormula object
     """
     candidate_space = None
+    ion_mode_int = 1 if mf.adduct.pos_mode else -1
     for i in range(len(mf.ms2_processed.mz_array)):
         # retrieve all indices of mass within tolerance
         this_ms2_tol = ms2_tol if not ppm else ms2_tol * mf.ms2_processed.mz_array[i] * 1e-6
@@ -785,10 +786,9 @@ def _assign_ms2_explanation(mf: MetaFeature, cf: CandidateFormula, pre_charged_a
         # retrieve all subformulas within tolerance
         this_subform_arr = subform_arr[idx_list, :]
         this_mass = mass_arr[idx_list]
-        ion_mode_int = 1 if mf.adduct.pos_mode else -1
 
         # dbe filter (DBE >= -1)
-        bool_arr_1 = _dbe_subform_filter(this_subform_arr)
+        bool_arr_1 = _dbe_subform_filter(this_subform_arr, -1.)
 
         # SENIOR rules filter, a soft version
         bool_arr_2 = _senior_subform_filter(this_subform_arr)
@@ -831,3 +831,81 @@ def _assign_ms2_explanation(mf: MetaFeature, cf: CandidateFormula, pre_charged_a
     candidate_form.ms1_isotope_similarity = cf.ms1_isotope_similarity  # copy ms1_isotope_similarity
 
     return candidate_form
+
+
+def assign_subformula(ms2_mz: List, precursor_formula: str, adduct: str,
+                      ms2_tol: float = 10, ppm: bool = True,
+                      dbe_cutoff: float = -1.0) -> Union[List[SubformulaResult], None]:
+    """
+    Assign subformulas to a given MS2 spectrum with a given precursor formula and adduct. Radical ions are considered.
+    :param ms2_mz: MS2 m/z list
+    :param precursor_formula: precursor formula string, uncharged
+    :param adduct: adduct string, e.g., [M+H]+
+    :param ms2_tol: MS2 tolerance
+    :param ppm: whether MS2 tolerance is in ppm
+    :param dbe_cutoff: float, DBE cutoff
+    :return: a list of SubformulaResult objects
+    """
+    assert len(ms2_mz) > 0, "MS2 m/z list is empty."
+
+    # convert precursor formula to array
+    form_arr = read_formula(precursor_formula)
+    if form_arr is None:
+        return None
+
+    # convert adduct to array
+    valid_adduct, pos_mode = check_adduct(adduct)
+    if not valid_adduct:
+        raise ValueError("Invalid adduct string.")
+    ion = Adduct(adduct, pos_mode)
+    ion_mode_int = 1 if pos_mode else -1
+
+    # enumerate all subformulas
+    pre_charged_arr = form_arr * ion.m + ion.net_formula.array
+    subform_arr = enumerate_subformula(pre_charged_arr)
+
+    # mono mass
+    mass_arr = _calc_subform_mass(subform_arr, ion.charge)
+
+    # assign ms2 explanation
+    out_list = []
+    for k, mz in enumerate(ms2_mz):
+        # retrieve all indices of mass within tolerance
+        this_ms2_tol = ms2_tol if not ppm else ms2_tol * mz * 1e-6
+        idx_list = np.where(abs(mz - mass_arr) <= this_ms2_tol)[0]
+
+        if len(idx_list) == 0:
+            out_list.append(SubformulaResult(k, []))
+            continue
+
+        # retrieve all subformulas within tolerance
+        this_subform_arr = subform_arr[idx_list, :]
+        this_mass = mass_arr[idx_list]
+
+        # dbe filter (DBE >= dbe_cutoff)
+        bool_arr_1 = _dbe_subform_filter(this_subform_arr, dbe_cutoff)
+
+        # SENIOR rules filter, a soft version
+        bool_arr_2 = _senior_subform_filter(this_subform_arr)
+
+        # valid subformula check
+        bool_arr_3 = _valid_subform_check(this_subform_arr, pre_charged_arr)
+
+        # combine filters
+        bool_arr = bool_arr_1 & bool_arr_2 & bool_arr_3
+        this_subform_arr = this_subform_arr[bool_arr, :]
+        this_mass = this_mass[bool_arr]
+
+        # if no valid subformula, skip
+        if this_subform_arr.shape[0] == 0:
+            out_list.append(SubformulaResult(k, []))
+            continue
+
+        # create SubformulaResult object
+        subform_list = []
+        for j in range(this_subform_arr.shape[0]):
+            form = Formula(this_subform_arr[j, :], ion_mode_int, this_mass[j])
+            subform_list.append(FormulaResult(form.__str__(), form.mass, mz))
+        out_list.append(SubformulaResult(k, subform_list))
+
+    return out_list

{msbuddy-0.2.0 → msbuddy-0.2.2}/msbuddy/load.py

@@ -161,10 +161,10 @@ def load_mgf(file_path) -> List[MetaFeature]:
                     key, value = _line.split('=', 1)
                     key, value = key.strip(), value.strip()
                     # if key (into all upper case) is 'PEPMASS', it is precursor mz
-                    if key.upper() == 'PEPMASS':
+                    if key.upper() in ['PEPMASS', 'PRECURSOR_MZ']:
                         precursor_mz = float(value)
                     # if key is 'TITLE', it is identifier
-                    elif key.upper() in ['TITLE', 'SCANS', 'SPECTRUMID']:
+                    elif key.upper() in ['TITLE', 'SPECTRUMID', 'SPECTRUM_ID']:
                         identifier = value.strip()
                     # if key is 'CHARGE' and charge is not set, it is charge
                     elif key.upper() == 'CHARGE':
@@ -295,12 +295,12 @@ def load_usi(usi_list: Union[str, List[str]],
             continue
     return data_list
 
-#
-# # test
-# if __name__ == '__main__':
-#     init_db()
 
+# test
+# if __name__ == '__main__':
     #################
+    # init_db()
+
     # import joblib
     # compile all these databases
 

{msbuddy-0.2.0 → msbuddy-0.2.2}/msbuddy/main.py

@@ -577,3 +577,4 @@ def _generate_candidate_formula(mf: MetaFeature, ps: MsbuddyConfig, global_dict)
     gen_candidate_formula(mf, ps.ppm, ps.ms1_tol, ps.ms2_tol, ps.db_mode, ps.ele_lower, ps.ele_upper,
                           ps.max_isotope_cnt, global_dict)
     return mf
+

{msbuddy-0.2.0 → msbuddy-0.2.2}/msbuddy/utils.py

@@ -25,7 +25,7 @@ alphabet = ["C", "H", "Br", "Cl", "F", "I", "K", "N", "Na", "O", "P", "S"]
 
 class FormulaResult:
     """
-    FormulaReturn class, for mass_to_formula function
+    FormulaResult class, for mass_to_formula function
     """
 
     def __init__(self, formula: str, mass: float, t_mass: float):
@@ -40,6 +40,21 @@ class FormulaResult:
         self.mass_error_ppm = self.mass_error / t_mass * 1e6
 
 
+class SubformulaResult:
+    """
+    SubformulaResult class, for assign_subformula function
+    """
+
+    def __init__(self, idx: int, subform_list: List[FormulaResult]):
+        """
+        Constructor
+        :param idx: index of the fragment ion
+        :param subform_list: list of FormulaResult objects
+        """
+        self.idx = idx
+        self.subform_list = subform_list
+
+
 def read_formula(form: str) -> np.array:
     """
     read neutral formula string and return a 12-dim array, return None if invalid

{msbuddy-0.2.0 → msbuddy-0.2.2}/msbuddy.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: msbuddy
-Version: 0.2.0
+Version: 0.2.2
 Summary: Bottom-up MS/MS interrogation-based molecular formula annotation for mass spectrometry data.
 Home-page: https://github.com/Philipbear/msbuddy
 Author: Shipei Xing
@@ -20,7 +20,13 @@ License-File: LICENSE
 [![docs](https://readthedocs.org/projects/msbuddy/badge/?version=latest)](https://msbuddy.readthedocs.io/en/latest/?badge=latest)
 [![Generic badge](https://img.shields.io/badge/msbuddy-mass_spec_tools-<COLOR>.svg)](https://github.com/Philipbear/msbuddy)
 
-`msbuddy` is developed for molecular formula annotation in mass spectrometry-based small molecule analysis.
+<p align="center">
+  <img src="https://github.com/Philipbear/msbuddy/blob/main/logo/logo.svg" alt="Sample Image" height="100"/>
+</p>
+
+
+`msbuddy` is developed for molecular formula analysis in MS-based metabolomics studies.
+
 `msbuddy` is able to provide formula annotations for queries:
   * with or without MS1 isotopic pattern 
   * with or without MS/MS spectra
@@ -30,7 +36,7 @@ The minimum requirement for a msbuddy query is a single _m/z_ value and its ioni
 
 **Official documentation**  is available at https://msbuddy.readthedocs.io/en/latest/.
 
-**msbuddy ver 0.2.0** is now released, with refined models and much improved performance. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
+**msbuddy ver 0.2.1** is now released. See [change log](https://github.com/Philipbear/msbuddy/blob/main/changelog.md) for details.
 
 ##  Citation
 > S. Xing et al. BUDDY: molecular formula discovery via bottom-up MS/MS interrogation. **Nature Methods** 2023. [DOI: 10.1038/s41592-023-01850-x](https://doi.org/10.1038/s41592-023-01850-x)
@@ -45,20 +51,21 @@ pip install msbuddy
 
 To install a specific version, see PyPI [release history](https://pypi.org/project/msbuddy/#history).
 
-## Python usage ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
+## Python usage 
 
+### Quick start ([docs](https://msbuddy.readthedocs.io/en/latest/quickstart.html))
 As a quick start, we here load a mgf file and annotate the molecular formula for each query.
 All the configurations are specified in the [`MsbuddyConfig`](https://msbuddy.readthedocs.io/en/latest/pyapi.html#msbuddy.MsbuddyConfig) class.
-**Parallel computing** is supported.
+**Parallel computing** is [supported](https://msbuddy.readthedocs.io/en/latest/quickstart.html).
 
 ```python
 from msbuddy import Msbuddy, MsbuddyConfig
 
 # instantiate a MsbuddyConfig object
-msb_config = MsbuddyConfig(ms_instr='orbitrap', # supported: "qtof", "orbitrap" and "fticr"
-                                                # highly recommended to specify
-                           halogen=False, # whether to consider halogen atoms FClBrI
-                           parallel=True, n_cpu=12)
+msb_config = MsbuddyConfig(# highly recommended to specify
+                           ms_instr='orbitrap',  # supported: "qtof", "orbitrap" and "fticr"
+                           # whether to consider halogen atoms FClBrI
+                           halogen=False)
 
 # instantiate a Msbuddy object
 msb_engine = Msbuddy(msb_config)
@@ -72,6 +79,8 @@ msb_engine.annotate_formula()
 # retrieve the annotation result summary
 result = msb_engine.get_summary()
 ```
+See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
+
 
 MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.1101/2020.05.09.086066v2) if they are publicly available:
 ```python
@@ -79,7 +88,14 @@ MS/MS spectra can be loaded via their [USIs](https://www.biorxiv.org/content/10.
 msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
-**msbuddy** also provides python APIs for generating molecular formulas from [a neutral mass](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mass_to_formula) or [a charged _m/z_ value](https://msbuddy.readthedocs.io/en/latest/pyapi.html#mz_to_formula).
+
+### Other API functions
+- generate molecular formulas from a neutral mass (mass_to_formula)
+- generate molecular formulas from a charged _m/z_ value (mz_to_formula)
+- predict formula feasibility using a deep learning model (predict_formula_feasibility)
+- enumerate all possible subformulas of a given precursor formula (enumerate_subform_arr)
+
+See more in the [documentation](https://msbuddy.readthedocs.io/en/latest/pyapi.html).
 
 ## Command-line tool ([docs](https://msbuddy.readthedocs.io/en/latest/cmdapi.html))
 
@@ -95,7 +111,7 @@ for a single USI string (from FT-ICR, for example):
 ```bash
 msbuddy -usi <usi_string> -output <output_folder> -ms fticr
 ```
-or a csv file ([templates files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
+or a csv file ([template files](https://github.com/Philipbear/msbuddy/tree/main/demo)) containing multiple USI strings (QTOF in this example):
 ```bash
 msbuddy -csv <csv_file> -output <output_folder> -ms qtof -details
 ```

{msbuddy-0.2.0 → msbuddy-0.2.2}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r", encoding="utf-8") as fh:
 
 setuptools.setup(
     name="msbuddy",
-    version="0.2.0",
+    version="0.2.2",
     author="Shipei Xing",
     author_email="s1xing@health.ucsd.edu",
     description="Bottom-up MS/MS interrogation-based molecular formula annotation for mass spectrometry data.",

msbuddy-0.2.0/msbuddy/__init__.py

@@ -1 +0,0 @@
-from .main import Msbuddy, MsbuddyConfig