msbuddy 0.3.2__tar.gz → 0.3.4__tar.gz

{msbuddy-0.3.2 → msbuddy-0.3.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: msbuddy
-Version: 0.3.2
+Version: 0.3.4
 Summary: Molecular formula annotation for MS-based small molecule analysis.
 Home-page: https://github.com/Philipbear/msbuddy
 Author: Shipei Xing
@@ -79,7 +79,7 @@ msb_engine.annotate_formula()
 result = msb_engine.get_summary()
 ```
 See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
-One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `SPECTRUMID`, or `SPECTRUM_ID`.
+One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `FEATURE_ID`, `SPECTRUMID`, or `SPECTRUM_ID`.
 
 
 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:
@@ -89,7 +89,7 @@ msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
 
-Here is a test case on [Google Colab](https://colab.research.google.com/drive/1jFnuzSssxMJtd29lrQwPOBs-qUzmKAic?usp=sharing).
+Here is a test case on [Google Colab](https://colab.research.google.com/drive/1oLIapoylGbh37ivhcAN-X4tK6l5kIIpf?usp=sharing).
 
 ### Other API functions
 - parse a formula string (e.g., C6H12O6, 2H2O, C5H7NO2.HCl) ([read_formula_str](https://msbuddy.readthedocs.io/en/latest/pyapi.html#read_formula_str))

{msbuddy-0.3.2 → msbuddy-0.3.4}/README.md

@@ -63,7 +63,7 @@ msb_engine.annotate_formula()
 result = msb_engine.get_summary()
 ```
 See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
-One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `SPECTRUMID`, or `SPECTRUM_ID`.
+One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `FEATURE_ID`, `SPECTRUMID`, or `SPECTRUM_ID`.
 
 
 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:
@@ -73,7 +73,7 @@ msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
 
-Here is a test case on [Google Colab](https://colab.research.google.com/drive/1jFnuzSssxMJtd29lrQwPOBs-qUzmKAic?usp=sharing).
+Here is a test case on [Google Colab](https://colab.research.google.com/drive/1oLIapoylGbh37ivhcAN-X4tK6l5kIIpf?usp=sharing).
 
 ### Other API functions
 - parse a formula string (e.g., C6H12O6, 2H2O, C5H7NO2.HCl) ([read_formula_str](https://msbuddy.readthedocs.io/en/latest/pyapi.html#read_formula_str))

{msbuddy-0.3.2 → msbuddy-0.3.4}/msbuddy/cand.py

@@ -235,11 +235,8 @@ def gen_candidate_formula(mf: MetaFeature, ppm: bool, ms1_tol: float, ms2_tol: f
         #     cf_list = _fill_in_db_existence(ms1_cand_form_ls, ms2_cand_form_ls,
         #                                     ms1_cand_form_str_ls, ms2_cand_form_str_ls)
 
-    # calculate neutral mass of the precursor ion
-    ion_mode = 1 if mf.adduct.pos_mode else -1
-    t_neutral_mass = (mf.mz * abs(mf.adduct.charge) - mf.adduct.net_formula.mass - ion_mode * 0.0005486) / mf.adduct.m
     # calculate mz error
-    cf_list = _calc_mz_error(cf_list, t_neutral_mass, ppm)
+    cf_list = _calc_mz_error(cf_list, mf.mz, ppm)
 
     # if MS1 isotope data is available and >1 iso peaks, calculate isotope similarity
     ms1_iso_available = mf.ms1_processed and len(mf.ms1_processed) > 1
@@ -247,27 +244,28 @@ def gen_candidate_formula(mf: MetaFeature, ppm: bool, ms1_tol: float, ms2_tol: f
         for k, cf in enumerate(cf_list):
             cf_list[k].ms1_isotope_similarity = _calc_ms1_iso_sim(cf, mf, max_isotope_cnt)
 
+    ms2_available = mf.ms2_processed
+
     # retain top candidate formulas
-    mf.candidate_formula_list = _retain_top_cand_form(cf_list, ms1_tol, ms1_iso_available, 400)
+    mf.candidate_formula_list = _retain_top_cand_form(cf_list, ms1_tol, ms1_iso_available, ms2_available, 350)
 
     return mf
 
 
-def _calc_mz_error(cf_list: List[CandidateFormula], t_neutral_mass: float,
-                   ppm: bool) -> List[CandidateFormula]:
+def _calc_mz_error(cf_list: List[CandidateFormula], precursor_mz: float, ppm: bool) -> List[CandidateFormula]:
     """
     Calculate mz error for each candidate formula.
     :param cf_list: list of candidate formulas
-    :param t_neutral_mass: theoretical neutral mass
+    :param precursor_mz: precursor mz
     :param ppm: whether to use ppm as the unit of tolerance
     :return: list of candidate formulas
     """
     if ppm:
         for k, cf in enumerate(cf_list):
-            cf_list[k].mz_error = (cf.formula.mass - t_neutral_mass) / t_neutral_mass * 1e6
+            cf_list[k].mz_error = (cf.charged_formula.mass - precursor_mz) / precursor_mz * 1e6
     else:
         for k, cf in enumerate(cf_list):
-            cf_list[k].mz_error = cf.formula.mass - t_neutral_mass
+            cf_list[k].mz_error = cf.charged_formula.mass - precursor_mz
     return cf_list
 
 
@@ -692,11 +690,13 @@ def _fill_in_db_existence(ms1_cand_list: List[CandidateFormula], ms2_cand_list:
 
 
 def _retain_top_cand_form(cf_list: List[CandidateFormula], ms1_tol: float, ms1_iso_available: bool,
-                          top_n: int) -> List[CandidateFormula]:
+                          ms2_available: bool, top_n: int) -> List[CandidateFormula]:
     """
     Retain top candidate formulas by explained MS2 intensity, then by mz difference.
     :param cf_list: candidate formula list
     :param ms1_tol: MS1 tolerance
+    :param ms1_iso_available: whether MS1 isotope data is available
+    :param ms2_available: whether MS2 data is available
     :param top_n: number of top candidate formulas to retain
     :return: retained candidate formula list
     """
@@ -708,9 +708,13 @@ def _retain_top_cand_form(cf_list: List[CandidateFormula], ms1_tol: float, ms1_i
             cf_list = sorted(cf_list, key=lambda x: (_calc_log_p_norm(x.mz_error, ms1_tol / 3) * 0.05 +
                                                      x.ms1_isotope_similarity), reverse=True)
         else:
-            # sort candidate list by mz_error_log_p + 4 * exp_frag_int_pnt
-            cf_list = sorted(cf_list, key=lambda x: (_calc_log_p_norm(x.mz_error, ms1_tol / 3) + 4 * x.exp_ms2_sum_int),
-                             reverse=True)
+            if ms2_available:
+                # sort candidate list by mz_error_log_p + 4 * exp_frag_int_pnt
+                cf_list = sorted(cf_list, key=lambda x: (_calc_log_p_norm(x.mz_error, ms1_tol / 3) + 4 * x.exp_ms2_sum_int),
+                                 reverse=True)
+            else:
+                # sort candidate list by abs mz_error, ascending
+                cf_list = sorted(cf_list, key=lambda x: abs(x.mz_error), reverse=False)
 
         return cf_list[:top_n]
 

{msbuddy-0.3.2 → msbuddy-0.3.4}/msbuddy/load.py

@@ -173,8 +173,8 @@ def load_mgf(file_path) -> List[MetaFeature]:
                     # if key (into all upper case) is 'PEPMASS', it is precursor mz
                     if key.upper() in ['PEPMASS', 'PRECURSOR_MZ']:
                         precursor_mz = float(value)
-                    # if key is 'TITLE', it is identifier
-                    elif key.upper() in ['TITLE', 'SPECTRUMID', 'SPECTRUM_ID']:
+                    # identifier
+                    elif key.upper() in ['TITLE', 'FEATURE_ID', 'SPECTRUMID', 'SPECTRUM_ID']:
                         identifier = value.strip()
                     # if key is 'CHARGE' and charge is not set, it is charge
                     elif key.upper() == 'CHARGE':

{msbuddy-0.3.2 → msbuddy-0.3.4}/msbuddy/main.py

@@ -232,7 +232,6 @@ class Msbuddy:
         :return: None. Update self.data
         """
 
-        #@timeout(self.config.timeout_secs)
         def _preprocess_and_gen_cand_nonparallel(meta_feature: MetaFeature, ps: MsbuddyConfig) -> MetaFeature:
             """
             a wrapper function for data preprocessing and candidate formula space generation
@@ -543,30 +542,31 @@ if __name__ == '__main__':
 
     start = time.time()
 
-    msb_config = MsbuddyConfig(ms_instr='orbitrap',  # supported: "qtof", "orbitrap" and "fticr"
-                               halogen=True, parallel=False, n_cpu=8)
+    msb_config = MsbuddyConfig(ms_instr='qtof',  # supported: "qtof", "orbitrap" and "fticr"
+                               halogen=True, parallel=True, n_cpu=10)
 
     msb_engine = Msbuddy(msb_config)
 
-    msb_engine.load_mgf('/Users/shipei/Documents/projects/msbuddy/demo/input_file.mgf')
+    # msb_engine.load_mgf('/Users/shipei/Documents/projects/msbuddy/demo/input_file.mgf')
 
-    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000085143_chagas_neg_orbi'  # 114
-    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000081463_tomato_pos'  # 265
-    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000086988_fecal_neg_orbi'  # 153
-    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000081981_AmericanGutProject'  # 168
-    # msb_engine.load_mgf(str(mgf_folder + '/ms1_ms2.mgf'))
+    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000085143_chagas_neg_orbi'  # 83
+    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000081463_tomato_pos'  # 194
+    # mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000086988_fecal_neg_orbi'  # 118
+    mgf_folder = '/Users/shipei/Documents/projects/msbuddy/results/lcms_datasets/MSV000081981_AmericanGutProject'  # 146
+    msb_engine.load_mgf(str(mgf_folder + '/ms1_ms2.mgf'))
 
     # cmd version
-    # msb_engine.annotate_formula_cmd(pathlib.Path(str(mgf_folder + '/msbuddy_output')), True)
+    msb_engine.annotate_formula_cmd(pathlib.Path(str(mgf_folder + '/msbuddy_output')), True)
 
     # annotate molecular formula
-    msb_engine.annotate_formula()
-    # retrieve the annotation result summary
-    result = msb_engine.get_summary()
+    # msb_engine.annotate_formula()
 
     end = time.time()
     print(end - start)
 
+    # retrieve the annotation result summary
+    result = msb_engine.get_summary()
+
     print(result)
     form_top1 = [r['formula_rank_1'] for r in result]
     form_est_fdr = [r['estimated_fdr'] for r in result]

{msbuddy-0.3.2 → msbuddy-0.3.4}/msbuddy.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: msbuddy
-Version: 0.3.2
+Version: 0.3.4
 Summary: Molecular formula annotation for MS-based small molecule analysis.
 Home-page: https://github.com/Philipbear/msbuddy
 Author: Shipei Xing
@@ -79,7 +79,7 @@ msb_engine.annotate_formula()
 result = msb_engine.get_summary()
 ```
 See [demo mgf file](https://github.com/Philipbear/msbuddy/tree/main/demo).
-One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `SPECTRUMID`, or `SPECTRUM_ID`.
+One of the following fields is recommended to be included, which will be used as the query identifier: `TITLE`, `FEATURE_ID`, `SPECTRUMID`, or `SPECTRUM_ID`.
 
 
 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:
@@ -89,7 +89,7 @@ msb_engine.load_usi(['mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740036',
                      'mzspec:GNPS:GNPS-LIBRARY:accession:CCMSLIB00003740037'])
 ```
 
-Here is a test case on [Google Colab](https://colab.research.google.com/drive/1jFnuzSssxMJtd29lrQwPOBs-qUzmKAic?usp=sharing).
+Here is a test case on [Google Colab](https://colab.research.google.com/drive/1oLIapoylGbh37ivhcAN-X4tK6l5kIIpf?usp=sharing).
 
 ### Other API functions
 - parse a formula string (e.g., C6H12O6, 2H2O, C5H7NO2.HCl) ([read_formula_str](https://msbuddy.readthedocs.io/en/latest/pyapi.html#read_formula_str))

{msbuddy-0.3.2 → msbuddy-0.3.4}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r", encoding="utf-8") as fh:
 
 setuptools.setup(
     name="msbuddy",
-    version="0.3.2",
+    version="0.3.4",
     author="Shipei Xing",
     author_email="s1xing@health.ucsd.edu",
     description="Molecular formula annotation for MS-based small molecule analysis.",