phykit 2.1.35__tar.gz → 2.1.38__tar.gz

{phykit-2.1.35 → phykit-2.1.38}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: phykit
-Version: 2.1.35
+Version: 2.1.38
 Home-page: https://github.com/jlsteenwyk/phykit
 Author: Jacob L. Steenwyk
 Author-email: jlsteenwyk@gmail.com

{phykit-2.1.35 → phykit-2.1.38}/phykit/cli_registry.py

@@ -157,6 +157,8 @@ ALIAS_TO_HANDLER: Dict[str, str] = {
     "etm": "evo_tempo_map",
     "disc_asym": "discordance_asymmetry",
     "da": "discordance_asymmetry",
+    "spec_disc": "spectral_discordance",
+    "sd": "spectral_discordance",
     # Helper aliases
     "create_concat": "create_concatenation_matrix",
     "cc": "create_concatenation_matrix",

{phykit-2.1.35 → phykit-2.1.38}/phykit/phykit.py

@@ -277,7 +277,10 @@ class Phykit:
                 treeness (alias: tness)
                     - reports treeness or stemminess, a measure of signal-to-
                       noise ratio in a phylogeny
-             
+                spectral_discordance (alias: spec_disc; sd)
+                    - PCA + spectral clustering of gene tree space via
+                      bipartition decomposition
+
                 Alignment- and tree-based commands
                 ==================================
                 saturation (alias: sat)
@@ -288,9 +291,13 @@ class Phykit:
                 """
             ),
         )
-        parser.add_argument("command", help=SUPPRESS)
+        parser.add_argument("command", nargs="?", default=None, help=SUPPRESS)
         args = parser.parse_args(sys.argv[1:2])
 
+        if args.command is None:
+            parser.print_help()
+            sys.exit(0)
+
         # if command is part of the possible commands (i.e., the long form
         # commands, run). Otherwise, assume it is an alias and look to the
         # run_alias function
@@ -302,7 +309,8 @@ class Phykit:
         except SystemExit:
             # Re-raise SystemExit as-is to preserve exit code
             raise
-        except NameError:
+        except NameError as e:
+            print(f"Error: {e}", file=sys.stderr)
             sys.exit(2)
 
     ## Aliases
@@ -491,7 +499,7 @@ class Phykit:
                 Aliases:
                   alignment_recoding, aln_recoding, recode
                 Command line interfaces: 
-                  bk_alignment_recoding, bk_aln_recoding, bk_recode
+                  pk_alignment_recoding, pk_aln_recoding, pk_recode
 
                 Usage:
                 phykit alignment_recoding <fasta> -c/--code <code> [--json]
@@ -1203,7 +1211,7 @@ class Phykit:
                 f"""\
                 {help_header}
 
-                Calculate the number and percentage of parismony
+                Calculate the number and percentage of parsimony
                 informative sites in an alignment.
 
                 The number of parsimony informative sites in an alignment
@@ -1850,13 +1858,13 @@ class Phykit:
 
                 Options
                 =====================================================
-                <tree_file_zero>            first argument after 
+                <tree_file_zero>            first argument after
                                             function name should be
-                                            an alignment file
+                                            a tree file
 
-                <tree_file_one>             first argument after 
+                <tree_file_one>             second argument after
                                             function name should be
-                                            an alignment file 
+                                            a tree file
 
                 -r/--reference              a tree to correct branch
                                             lengths by in the two input
@@ -4511,7 +4519,7 @@ class Phykit:
                                             a tree file
 
                 -v/--verbose                optional argument to print
-                                            all internal branch lengths
+                                            all terminal branch lengths
 
                 --json                      optional argument to output
                                             results as JSON
@@ -5000,6 +5008,95 @@ class Phykit:
         _add_json_argument(parser)
         _run_service(parser, argv, DiscordanceAsymmetry)
 
+    @staticmethod
+    def spectral_discordance(argv):
+        parser = _new_parser(
+            description=textwrap.dedent(
+                f"""\
+                {help_header}
+
+                Spectral discordance decomposition — decompose gene tree
+                space via PCA on a bipartition presence/absence (or
+                branch-length) matrix, with spectral clustering and
+                automatic cluster detection via the eigengap heuristic.
+
+                Each gene tree is encoded as a vector over the union of
+                all bipartitions observed across gene trees. PCA reveals
+                the axes of topological variation, with loading vectors
+                identifying which bipartitions drive each PC. Spectral
+                clustering groups genes sharing similar topologies.
+
+                Two metrics are available:
+                - nrf (default): binary presence/absence (normalized RF)
+                - wrf: branch-length weighted
+
+                Aliases:
+                  spectral_discordance, spec_disc, sd
+                Command line interfaces:
+                  pk_spectral_discordance, pk_spec_disc, pk_sd
+
+                Usage:
+                phykit spectral_discordance -g <gene_trees> [-t <tree>] [--metric nrf|wrf] [--clusters K] [--n-pcs N] [--top-loadings N] [--plot <prefix>] [--json]
+
+                Options
+                =====================================================
+                -g/--gene-trees             file of gene trees (one
+                                            Newick per line, or file
+                                            of filenames)
+
+                -t/--tree                   species tree (optional; flags
+                                            species-tree bipartitions in
+                                            loading output)
+
+                --metric                    distance metric: nrf or wrf
+                                            (default: nrf)
+
+                --clusters                  override auto-detected K
+
+                --n-pcs                     number of PCs to report
+                                            (default: min(10, G-1))
+
+                --top-loadings              top bipartitions per PC
+                                            (default: 5)
+
+                --plot                      output prefix for plots
+                                            (generates _scatter.png and
+                                            _eigengap.png)
+
+                --json                      output results as JSON
+                """
+            ),
+        )
+        parser.add_argument(
+            "-g", "--gene-trees", type=str, required=True, help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "-t", "--tree", type=str, required=False, default=None,
+            help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "--metric", type=str, required=False, default="nrf",
+            choices=["nrf", "wrf"], help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "--clusters", type=int, required=False, default=None,
+            help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "--n-pcs", type=int, required=False, default=None,
+            help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "--top-loadings", type=int, required=False, default=5,
+            help=SUPPRESS, metavar=""
+        )
+        parser.add_argument(
+            "--plot", type=str, required=False, default=None,
+            help=SUPPRESS, metavar=""
+        )
+        _add_json_argument(parser)
+        _run_service(parser, argv, SpectralDiscordance)
+
     ### Helper commands
     @staticmethod
     def create_concatenation_matrix(argv):
@@ -5453,3 +5550,15 @@ def create_concatenation_matrix(argv=None):
 
 def thread_dna(argv=None):
     Phykit.thread_dna(sys.argv[1:])
+
+
+def evo_tempo_map(argv=None):
+    Phykit.evo_tempo_map(sys.argv[1:])
+
+
+def discordance_asymmetry(argv=None):
+    Phykit.discordance_asymmetry(sys.argv[1:])
+
+
+def spectral_discordance(argv=None):
+    Phykit.spectral_discordance(sys.argv[1:])

{phykit-2.1.35 → phykit-2.1.38}/phykit/service_factories.py

@@ -98,6 +98,7 @@ Treeness = _LazyServiceFactory("phykit.services.tree.treeness", "Treeness")
 TreenessOverRCV = _LazyServiceFactory("phykit.services.tree.treeness_over_rcv", "TreenessOverRCV")
 EvoTempoMap = _LazyServiceFactory("phykit.services.tree.evo_tempo_map", "EvoTempoMap")
 DiscordanceAsymmetry = _LazyServiceFactory("phykit.services.tree.discordance_asymmetry", "DiscordanceAsymmetry")
+SpectralDiscordance = _LazyServiceFactory("phykit.services.tree.spectral_discordance", "SpectralDiscordance")
 
 SERVICE_FACTORIES: Dict[str, _LazyServiceFactory] = {
     name: value