Merge pull request #81 from faridrashidi/develop

release v0.0.19

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.0.18
+current_version = 0.0.19
 commit = True
 message = [skip ci] {current_version} → {new_version}
 tag = False

.github/ISSUE_TEMPLATE/bug_report.md

@@ -15,6 +15,7 @@ assignees: ''
 ```python
 ...
 ```
+
 ### Traceback
 <!-- If applicable, replace `...` with an error traceback below -->
 <details>

docs/source/references.bib

@@ -474,3 +474,12 @@ @article{Bourque
   title         = {{The Bourque distances for mutation trees of cancers}},
   journal       = {Algorithms for Molecular Biology}
 }
+@article{GRMT,
+  doi           = {10.3389/fgene.2021.692964},
+  year          = 2021,
+  month         = jun,
+  publisher     = {Frontiers Media {SA}},
+  volume        = {12},
+  author        = {Zhenhua Yu and Huidong Liu and Fang Du and Xiaofen Tang},
+  title         = {{{GRMT}: Generative Reconstruction of Mutation Tree From Scratch Using Single-Cell Sequencing Data}}
+}

docs/source/release_notes.rst

@@ -7,6 +7,16 @@ Release Notes
 =============
 
 
+Version 0.0.19 :small:`October 18, 2021`
+----------------------------------------
+
+This version includes:
+
+    - Add GRMT and SPhyR as new solvers.
+    - Update the consensus to new algorithm.
+    - Fix some bugs.
+
+
 Version 0.0.18 :small:`October 13, 2021`
 ----------------------------------------
 

requirements.txt

@@ -20,4 +20,5 @@ pyyaml
 scikit_learn
 scipy
 seaborn
+termcolor
 tqdm

setup.cfg

@@ -23,3 +23,23 @@ ignore =
     C400, # C400 Unnecessary generator - rewrite as a list comprehension
 [tool:pytest]
 filterwarnings = ignore::DeprecationWarning
+[coverage:run]
+omit =
+    ./trisicell/external/*
+    # unfinished
+    ./trisicell/tl/solver/_cardelino.py
+    ./trisicell/tl/solver/_cellphy.py
+    ./trisicell/tl/solver/_dendro.py
+    ./trisicell/tl/solver/_gpps.py
+    ./trisicell/tl/solver/_grmt.py
+    ./trisicell/tl/solver/_sasc.py
+    ./trisicell/tl/solver/_sbm.py
+    ./trisicell/tl/solver/_scelestial.py
+    ./trisicell/tl/solver/_siclonefit.py
+    ./trisicell/tl/solver/_sphyr.py
+    ./trisicell/commands/_grmt.py
+    ./trisicell/commands/_sphyr.py
+    # gurobi
+    ./trisicell/pp/_bifiltering.py
+    # r
+    ./trisicell/tl/cna/_cna.py

setup.py

@@ -12,7 +12,7 @@
     __author__ = ", ".join(["Farid Rashidi"])
     __maintainer__ = ", ".join(["Farid Rashidi"])
     __email__ = ", ".join(["[email protected]"])
-    __version__ = "0.0.18"
+    __version__ = "0.0.19"
 
 if platform == "linux" or platform == "linux2":
     os.environ["CC"] = "g++"

tests/test_datasets.py

@@ -21,7 +21,6 @@ def test_simulate_2(self):
         assert not tsc.ul.is_conflict_free_gusfield(df_noisy)
 
     def test_load_datasets(self):
-        # TODO: complete
         adata = tsc.datasets.example()
         assert adata.shape == (83, 452)
         adata = tsc.datasets.test()
@@ -47,7 +46,7 @@ def test_load_datasets(self):
         adata = tsc.datasets.erbc()
         assert adata.shape == (47, 40)
         adata = tsc.datasets.high_grade_serous_ovarian_cancer_3celllines()
-        assert adata.shape == (891, 13666)
+        assert adata.shape == (891, 14068)
         adata = tsc.datasets.melanoma20()
         assert adata.shape == (20, 2367)
         adata = tsc.datasets.muscle_invasive_bladder()

tests/test_io.py

@@ -0,0 +1,11 @@
+import trisicell as tsc
+
+
+class TestIO:
+    def test_read_newick(self):
+        f1 = tsc.ul.get_file("trisicell.datasets/test/T00.nwk")
+        f2 = tsc.ul.get_file("trisicell.datasets/test/T06.nwk")
+        df1 = tsc.io.read(f1)
+        df2 = tsc.io.read(f2)
+        assert df1.shape == (185, 368)
+        assert df2.shape == (185, 368)

tests/test_preprocessing.py → tests/test_pp.py

@@ -37,3 +37,7 @@ def test_readcount(self):
         tsc.pp.filter_mut_mutant_must_present_in_at_least(adata, min_cells=2)
         tsc.pp.statistics(adata)
         assert adata.shape == (83, 267)
+        tsc.pp.group_obs_apply_func(adata, group_key="group")
+        tsc.pp.remove_cell_by_list(adata, ["C15_1"])
+        tsc.pp.keep_cell_by_list(adata, ["C15_2", "C15_3"])
+        assert adata.shape == (2, 267)

tests/test_consensus.py → tests/test_tl_consensus.py

@@ -13,7 +13,7 @@ def test_consensus_1(self):
         sc1 = tsc.io.read(f1)
         sc2 = tsc.io.read(f2)
         final_tree = tsc.tl.consensus(sc1, sc2)
-        assert len(final_tree.nodes) == 18
+        assert len(final_tree.nodes) == 19
 
     def test_consensus_2(self):
         # result in recomb.fig1c
@@ -31,7 +31,7 @@ def test_consensus_3(self):
         sc1 = tsc.io.read(f1)
         sc2 = tsc.io.read(f2)
         final_tree = tsc.tl.consensus(sc1, sc2)
-        assert len(final_tree.nodes) == 33
+        assert len(final_tree.nodes) == 34
 
     def test_consensus_4(self):
         # result in biorxiv.fig3f
@@ -54,9 +54,7 @@ def test_consensus_day_1(self):
         sc2 = tsc.io.read(f2)
         tris_tree = tsc.tl.consensus(sc1, sc2)
         day_tree = tsc.tl.consensus_day(sc1, sc2)
-        assert not nx.is_isomorphic(
-            tris_tree, day_tree
-        )  # because of nonoverlapping mutations
+        assert nx.is_isomorphic(tris_tree, day_tree)
 
     def test_consensus_day_2(self):
         # result in recomb.fig1c
@@ -76,9 +74,7 @@ def test_consensus_day_3(self):
         sc2 = tsc.io.read(f2)
         tris_tree = tsc.tl.consensus(sc1, sc2)
         day_tree = tsc.tl.consensus_day(sc1, sc2)
-        assert not nx.is_isomorphic(
-            tris_tree, day_tree
-        )  # because of nonoverlapping mutations in private of C21
+        assert nx.is_isomorphic(tris_tree, day_tree)
 
     def test_consensus_day_4(self):
         # result in biorxiv.fig3f

tests/test_solvers.py → tests/test_tl_solvers.py

@@ -32,11 +32,6 @@ def test_phiscsb(self):
         df_out = tsc.tl.phiscsb(self.df_in, alpha=0.0000001, beta=0.1)
         assert tsc.ul.is_conflict_free_gusfield(df_out)
 
-    @skip_gurobi
-    def test_phiscsi(self):
-        df_out = tsc.tl.phiscsi(self.df_in, alpha=0.0000001, beta=0.1)
-        assert tsc.ul.is_conflict_free_gusfield(df_out)
-
     def test_huntress_both(self):
         df_out = tsc.tl.huntress(self.df_in, alpha=0.0000001, beta=0.1, kind="both")
         assert tsc.ul.is_conflict_free_gusfield(df_out)
@@ -51,6 +46,11 @@ def test_onconem(self):
         is_cf = tsc.ul.is_conflict_free_gusfield(df_out)
         assert is_cf
 
+    @skip_gurobi
+    def test_phiscsi(self):
+        df_out = tsc.tl.phiscsi(self.df_in, alpha=0.0000001, beta=0.1)
+        assert tsc.ul.is_conflict_free_gusfield(df_out)
+
     @skip_gurobi
     def test_phiscsi_bulk_1(self):
         adata = tsc.datasets.acute_lymphocytic_leukemia2()

tests/test_solvers_tmp.py → tests/test_tl_solvers_tmp.py

@@ -18,6 +18,12 @@ def test_titch_algorithm(self):
         tsc.tl.fitch(tree)
         assert True
 
+    def test_rscistree(self):
+        adata = tsc.datasets.colorectal2(readcount=True)
+        df_out = tsc.tl.rscistree(adata, mode="haploid")
+        is_cf = tsc.ul.is_conflict_free_gusfield(df_out)
+        assert is_cf
+
     def test_iscistree(self):
         df_in = tsc.datasets.test()
         df_out = tsc.tl.iscistree(df_in, alpha=0.0000001, beta=0.1)
@@ -59,10 +65,3 @@ def test_cardelino(self):
         adata = tsc.datasets.example()
         tsc.tl.cardelino(adata, mode="free", n_clones=11)
         assert True
-
-    @pytest.mark.skip(reason="Cause make: *** Error!")
-    def test_rscistree(self):
-        adata = tsc.datasets.colorectal2(readcount=True)
-        df_out = tsc.tl.rscistree(adata, mode="haploid")
-        is_cf = tsc.ul.is_conflict_free_gusfield(df_out)
-        assert is_cf

tests/test_utils.py → tests/test_ul.py

@@ -81,3 +81,21 @@ def test_cells_muts_rooted_at(self):
 
         res = tsc.ul.muts_rooted_at(tree, "[8]")
         assert res.shape[0] == 51
+
+    def test_general(self):
+        # tsc.ul.executable("kDPFC", "SPhyR")
+        file = tsc.ul.get_file("trisicell.datasets/test/fp_0-fn_0-na_0.ground.CFMatrix")
+        tsc.ul.dir_base(file)
+        tsc.ul.dirbase(file)
+        tsc.ul.get_param(
+            "simNo_2-s_7-m_20-h_1-minVAF_0.1-ISAV_0-n_10-fp_0-fn_0.1-na_0-d_0-l_1000000"
+            ".SC"
+        )
+
+        @tsc.ul.timeit
+        def _test():
+            return None
+
+        _test()
+
+        assert True

trisicell/__init__.py

@@ -7,5 +7,5 @@
 __author__ = ", ".join(["Farid Rashidi"])
 __maintainer__ = ", ".join(["Farid Rashidi"])
 __email__ = ", ".join(["[email protected]"])
-__version__ = "0.0.18"
+__version__ = "0.0.19"
 __all__ = (datasets, io, logg, pl, pp, settings, tl, ul)

trisicell/commands/_grmt.py

@@ -0,0 +1,52 @@
+import os
+
+import click
+
+import trisicell as tsc
+
+
+@click.command(short_help="Run GRMT.")
+@click.argument(
+    "genotype_file",
+    required=True,
+    type=click.Path(
+        exists=True, file_okay=True, dir_okay=False, readable=True, resolve_path=True
+    ),
+)
+@click.argument(
+    "alpha",
+    required=True,
+    type=float,
+)
+@click.argument(
+    "beta",
+    required=True,
+    type=float,
+)
+@click.option(
+    "--n_threads",
+    "-p",
+    default=1,
+    type=int,
+    show_default=True,
+    help="Number of threads.",
+)
+def grmt(genotype_file, alpha, beta, n_threads):
+    """GRMT.
+
+    Generative Reconstruction of Mutation Tree From Scratch Using Single-Cell
+    Sequencing Data :cite:`GRMT`.
+
+    trisicell grmt input.SC 0.0001 0.1
+    """
+
+    outfile = os.path.splitext(genotype_file)[0]
+
+    tsc.settings.verbosity = "info"
+    tsc.settings.logfile = f"{outfile}.grmt.log"
+
+    df_in = tsc.io.read(genotype_file)
+    df_out = tsc.tl.grmt(df_in, alpha=alpha, beta=beta, n_threads=n_threads)
+    tsc.io.write(df_out, f"{outfile}.grmt.CFMatrix")
+
+    return None

trisicell/commands/_phiscs.py

@@ -69,7 +69,7 @@ def phiscsb(genotype_file, alpha, beta):
     default=86400,
     type=int,
     show_default=True,
-    help="Timelimit of the program (in second).",
+    help="Time limit of the program (in second).",
 )
 @click.option(
     "--n_threads",

trisicell/commands/_score.py

@@ -33,7 +33,14 @@ def score(ground_file, inferred_file):
     dl = tsc.tl.dl(df_g, df_s)
     mltd = tsc.tl.mltd(df_g, df_s)["normalized_similarity"]
     tpted = tsc.tl.tpted(df_g, df_s)
-
-    tsc.logg.info(f"AD={ad:0.4f}\nDL={dl:0.4f}\nMLTSM={mltd:0.4f}\nTPTED={tpted:0.4f}")
+    caset = tsc.tl.caset(df_g, df_s)
+    disc = tsc.tl.disc(df_g, df_s)
+    mp3 = tsc.tl.mp3(df_g, df_s)
+    rf = tsc.tl.rf(df_g, df_s)
+
+    tsc.logg.info(
+        f"AD={ad:0.4f}\nDL={dl:0.4f}\nMLTSM={mltd:0.4f}\nTPTED={tpted:0.4f}\n"
+        f"CASet={caset:0.4f}\nDISC={disc:0.4f}\nMP3={mp3:0.4f}\nRF={rf:0.4f}"
+    )
 
     return None

trisicell/commands/_sphyr.py

@@ -0,0 +1,62 @@
+import os
+
+import click
+
+import trisicell as tsc
+
+
+@click.command(short_help="Run SPhyR.")
+@click.argument(
+    "genotype_file",
+    required=True,
+    type=click.Path(
+        exists=True, file_okay=True, dir_okay=False, readable=True, resolve_path=True
+    ),
+)
+@click.argument(
+    "alpha",
+    required=True,
+    type=float,
+)
+@click.argument(
+    "beta",
+    required=True,
+    type=float,
+)
+@click.option(
+    "--n_restarts",
+    "-r",
+    default=10,
+    type=int,
+    show_default=True,
+    help="Number of restarts.",
+)
+@click.option(
+    "--n_threads",
+    "-p",
+    default=1,
+    type=int,
+    show_default=True,
+    help="Number of threads.",
+)
+def sphyr(genotype_file, alpha, beta, n_restarts, n_threads):
+    """SPhyR.
+
+    Tumor phylogeny estimation from single-cell sequencing data under loss and error
+    :cite:`SPhyR`.
+
+    trisicell sphyr input.SC 0.0001 0.1
+    """
+
+    outfile = os.path.splitext(genotype_file)[0]
+
+    tsc.settings.verbosity = "info"
+    tsc.settings.logfile = f"{outfile}.sphyr.log"
+
+    df_in = tsc.io.read(genotype_file)
+    df_out = tsc.tl.sphyr(
+        df_in, alpha=alpha, beta=beta, n_restarts=n_restarts, n_threads=n_threads
+    )
+    tsc.io.write(df_out, f"{outfile}.sphyr.CFMatrix")
+
+    return None