Release v1.2.0 (#18)

* Add "positive" flag for correlation analysis and update pandas data fra… (#15)

* Add check box to GUI for "filtering" positive correlations

* Update CLI for neutral losses

* Update GitHub actions and configs (#17)

* Update README.rst (Minor)

---------

Co-authored-by: Wanchang Lin <wanchang_lin@outlook.com>

.github/workflows/build-test.yml

@@ -9,7 +9,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest, windows-latest, macos-latest]
-        python-version: [3.7, 3.8]
+        python-version: ['3.8', '3.9', '3.10']
 
     env:
       OS: ${{ matrix.os }}
@@ -29,7 +29,7 @@ jobs:
         run: |
 
           python --version
-          conda env update --file environment.yml --name base
+          conda env update --file environment.yml --name __setup_conda
 
       - name: Lint with flake8
         run: |
@@ -44,15 +44,16 @@ jobs:
 
       - name: Test with pytest-cov
         run: |
-
-          python setup.py install
-          beamspy --help
-
           conda install pytest codecov pytest-cov -c conda-forge
+          
+          python -m pip install --no-deps -e .
+          beamspy --help
+          
           pytest --cov ./ --cov-config=.coveragerc --cov-report=xml
 
       - name: Upload code coverage to codecov
         uses: codecov/codecov-action@v1
+        if: matrix.os == 'ubuntu-latest'
         with:
           flags: unittests
           env_vars: OS,PYTHON

README.rst

@@ -1,6 +1,6 @@
 BEAMSpy - Birmingham mEtabolite Annotation for Mass Spectrometry (Python package)
 ==================================================================================
-|Version| |Py versions| |Git| |Bioconda| |Build Status| |Build Status (AppVeyor)| |License| |RTD doc| |codecov| |mybinder|
+|Version| |Py versions| |Git| |Bioconda| |Build Status| |License| |RTD doc| |codecov| |mybinder|
 
 BEAMSpy (Birmingham mEtabolite Annotation for Mass Spectrometry) is a Python package that includes several automated and
 seamless computational modules that are applied to putatively annotate metabolites detected in untargeted ultra (high)
@@ -82,9 +82,6 @@ Released under the GNU General Public License v3.0 (see `LICENSE <https://github
 .. |Build Status| image:: https://github.com/computational-metabolomics/beamspy/workflows/beamspy/badge.svg
    :target: https://github.com/computational-metabolomics/beamspy/actions
 
-.. |Build Status (AppVeyor)| image:: https://img.shields.io/appveyor/ci/RJMW/beamspy.svg?style=flat&maxAge=3600&label=AppVeyor
-   :target: https://ci.appveyor.com/project/RJMW/beamspy
-
 .. |Py versions| image:: https://img.shields.io/pypi/pyversions/beamspy.svg?style=flat&maxAge=3600
    :target: https://pypi.python.org/pypi/beamspy/
 

beamspy/__init__.py

@@ -1,4 +1,4 @@
 __author__ = 'r.j.weber@bham.ac.uk'
 __credits__ = 'r.j.weber@bham.ac.uk'
-__version__ = '1.1.0'
-__license__ = 'GPLv3'
+__version__ = '1.2.0'
+__license__ = 'GPLv3'

beamspy/__main__.py

@@ -71,6 +71,9 @@ def main():
     parser_gf.add_argument('-p', '--pvalue-threshold', default=0.01, type=float, required=True,
                            help="Threshold for p-value.")
 
+    parser_gf.add_argument('-o', '--positive', action="store_true",
+                           help="Use positive correlation only otherwise use both positive and negative correlation.")
+
     parser_gf.add_argument('-g', '--gml-file', type=str, required=True,
                            help="Write graph to GraphML format.")
 
@@ -237,7 +240,7 @@ def main():
         df = in_out.combine_peaklist_matrix(args.peaklist, args.intensity_matrix)
         graph = grouping.group_features(df, db_out=args.db, max_rt_diff=args.max_rt_diff,
                                         coeff_thres=args.coeff_threshold, pvalue_thres=args.pvalue_threshold,
-                                        method=args.method, ncpus=args.ncpus)
+                                        method=args.method, positive=args.positive, ncpus=args.ncpus)
         nx.write_gml(graph, str(args.gml_file))
 
     if args.step == "annotate-peak-patterns":
@@ -274,7 +277,7 @@ def main():
                 path = 'data/neutral_losses.txt'
                 p = os.path.join(os.path.dirname(os.path.abspath(__file__)), path)
                 lib = in_out.read_neutral_losses(p)
-            annotation.neutral_losses(inp, db_out=args.db, ppm=args.ppm, lib=lib)
+            annotation.annotate_neutral_losses(inp, db_out=args.db, ppm=args.ppm, lib=lib)
 
         if args.oligomers:
             if args.adducts_library:

beamspy/annotation.py

@@ -1615,7 +1615,8 @@ def summary(df, db, single_row=False, single_column=False, convert_rt=None, ndig
 
         records = [(str(record[0]), str(record[1])) for record in cursor.fetchall()]
 
-        G = nx.OrderedDiGraph()
+        # G = nx.OrderedDiGraph()  # networkx version < 3.0
+        G = nx.DiGraph()
         G.add_edges_from(records)
 
         graphs = list(G.subgraph(c) for c in nx.weakly_connected_components(G))

beamspy/data/adducts.txt

@@ -1 +1,9 @@
-label	exact_mass	charge	ion_mode[M+H]+	1.007276	1	pos[M+Na]+	22.989221	1	pos[M+K]+	38.963158	1	pos[M-H]-	-1.007276	1	neg[M+Cl]-	34.969401	1	neg[M+Na-2H]-	20.974668	1	neg[M+K-2H]-	36.948605	1	neg[M+Hac-H]-	59.013853	1	neg
+label	exact_mass	charge	ion_mode
+[M+H]+	1.007276	1	pos
+[M+Na]+	22.989221	1	pos
+[M+K]+	38.963158	1	pos
+[M-H]-	-1.007276	1	neg
+[M+Cl]-	34.969401	1	neg
+[M+Na-2H]-	20.974668	1	neg
+[M+K-2H]-	36.948605	1	neg
+[M+Hac-H]-	59.013853	1	neg

beamspy/data/isotopes.txt

@@ -1 +1,6 @@
-label_x	label_y	mass_difference	abundance_x	abundance_y	ion_mode	chargeC	(13C)	1.003355	98.93	1.07	both	1C	(13C)	0.5016775	98.93	1.07	both	2S	(34S)	1.995796	94.99	4.25	both	1K	(41K)	1.998119	93.25	6.73	pos	1Cl	(37Cl)	1.99705	75.76	24.24	neg	1
+label_x	label_y	mass_difference	abundance_x	abundance_y	ion_mode	charge
+C	(13C)	1.003355	98.93	1.07	both	1
+C	(13C)	0.5016775	98.93	1.07	both	2
+S	(34S)	1.995796	94.99	4.25	both	1
+K	(41K)	1.998119	93.25	6.73	pos	1
+Cl	(37Cl)	1.99705	75.76	24.24	neg	1

beamspy/data/multiple_charged_ions.txt

@@ -1 +1,5 @@
-label	exact_mass	charge	ion_mode[M+H]+	1.007276	1	pos[M+Na]+	22.989221	1	pos[M+2H]2+	1.007276	2	pos[M+H+Na]2+	11.9982485	2	pos
+label	exact_mass	charge	ion_mode
+[M+H]+	1.007276	1	pos
+[M+Na]+	22.989221	1	pos
+[M+2H]2+	1.007276	2	pos
+[M+H+Na]2+	11.9982485	2	pos

beamspy/data/neutral_losses.txt

@@ -1 +1,3 @@
-label	mass_differenceH2O	18.010565CO	27.994915
+label	mass_difference
+H2O	18.010565
+CO	27.994915

beamspy/grouping.py

@@ -6,7 +6,7 @@
 import networkx as nx
 
 
-def group_features(df, db_out, max_rt_diff=5.0, coeff_thres=0.7, pvalue_thres=1.0, method="pearson", block=5000, ncpus=None):
+def group_features(df, db_out, max_rt_diff=5.0, coeff_thres=0.7, pvalue_thres=1.0, method="pearson", positive=True, block=5000, ncpus=None):
 
     conn = sqlite3.connect(db_out)
     cursor = conn.cursor()
@@ -25,7 +25,7 @@ def group_features(df, db_out, max_rt_diff=5.0, coeff_thres=0.7, pvalue_thres=1.
                    mz_diff REAL DEFAULT NULL,                 
                    PRIMARY KEY (peak_id_a, peak_id_b));""")
 
-    df_coeffs = statistics.correlation_coefficients(df, max_rt_diff, coeff_thres, pvalue_thres, method, block, ncpus)
+    df_coeffs = statistics.correlation_coefficients(df, max_rt_diff, coeff_thres, pvalue_thres, method, positive, block, ncpus)
     graph = statistics.correlation_graphs(df_coeffs, df)
     sub_graphs = list(graph.subgraph(c) for c in nx.weakly_connected_components(graph))
     for i in range(len(sub_graphs)):

beamspy/gui.py

@@ -188,6 +188,7 @@ def group_features(self):
             self.label_grouping_ncpus.setEnabled(False)
             self.doubleSpinBox_ncpus.setEnabled(False)
             self.doubleSpinBox_block.setEnabled(False)
+            self.checkBox_positive_correlation.setEnabled(True)
         else:
             self.label_max_rt.setEnabled(True)
             self.doubleSpinBox_max_rt.setEnabled(True)
@@ -201,6 +202,7 @@ def group_features(self):
             self.label_grouping_ncpus.setEnabled(True)
             self.doubleSpinBox_ncpus.setEnabled(True)
             self.doubleSpinBox_block.setEnabled(True)
+            self.checkBox_positive_correlation.setEnabled(True)
         self.source_graph_file()
 
     def annotate_peak_patterns(self):
@@ -355,6 +357,7 @@ def run(self):
                                             coeff_thres=self.doubleSpinBox_coefficent.value(),
                                             pvalue_thres=self.doubleSpinBox_p_value.value(),
                                             method=method,
+                                            positive=self.checkBox_positive_correlation.isChecked(),
                                             block=int(self.doubleSpinBox_block.value()),
                                             ncpus=int(self.doubleSpinBox_ncpus.value()))
             nx.write_gml(graph, str(self.lineEdit_graph.text()))

beamspy/in_out.py

@@ -189,9 +189,9 @@ def combine_peaklist_matrix(fn_peaklist, fn_matrix, separator="\t", median_inten
 
     if mapping["intensity"] not in df_peaklist.columns:
         if median_intensity:
-            df_peaklist["intensity"] = pd.Series(df_matrix.median(axis=1, skipna=True), index=df_matrix.index)
+            df_peaklist["intensity"] = pd.Series(df_matrix.median(axis=1, skipna=True, numeric_only=True), index=df_matrix.index)
         else:
-            df_peaklist["intensity"] = pd.Series(df_matrix.mean(axis=1, skipna=True), index=df_matrix.index)
+            df_peaklist["intensity"] = pd.Series(df_matrix.mean(axis=1, skipna=True, numeric_only=True), index=df_matrix.index)
 
     if len(df_peaklist[mapping["name"]].unique()) != len(df_peaklist[mapping["name"]]):
         raise ValueError("Peaklist: Values column '{}' are not unique".format(mapping["name"]))