MRG: try using sourmash_utils functions. (#15)

* try using new sourmash_utils stuff * add sourmash_utils as a dep * upd sourmash utils usage * refactor around sourmash_utils * bump version * fix CI
dib-lab · Jun 11, 2024 · a0e2970 · a0e2970
1 parent 8b42d09
commit a0e2970
Show file tree

Hide file tree

Showing 6 changed files with 47 additions and 58 deletions.
diff --git a/.github/workflows/build-test.yml b/.github/workflows/build-test.yml
@@ -47,4 +47,4 @@ jobs:
 
     - name: build examples
       shell: bash -l {0}
-      run: make cleanrun
+      run: make cleanall
diff --git a/Makefile b/Makefile
@@ -1,4 +1,4 @@
-.PHONY: dist test_workflow
+.PHONY: dist test_workflow clean cleanall
 
 all: test_workflow
 
@@ -14,7 +14,9 @@ install:
 test_workflow:
 	cd test_workflow && make
 
-cleanrun:
+clean: cd test_workflow && make clean
+
+cleanall:
 	cd test_workflow && make cleanall
 
 dist:

diff --git a/pyproject.toml b/pyproject.toml
@@ -3,13 +3,13 @@ name = "sourmash_plugin_pangenomics"
 description = "sourmash plugin to do pangenomics."
 readme = "README.md"
 requires-python = ">=3.10"
-version = "0.2"
+version = "0.2.1"
 authors = [
   {name = "Colton Baumler", email = "[email protected]"},
   {name = "Titus Brown", email = "[email protected]"},
 ]
 
-dependencies = ["sourmash>=4.8.8,<5"]
+dependencies = ["sourmash>=4.8.8,<5", "sourmash_utils>=0.2"]
 
 [metadata]
 license = { text = "BSD 3-Clause License" }

diff --git a/src/sourmash_plugin_pangenomics.py b/src/sourmash_plugin_pangenomics.py
@@ -15,6 +15,7 @@
 import pprint
 
 import sourmash
+import sourmash_utils
 from sourmash import sourmash_args
 from sourmash.tax import tax_utils
 from sourmash.logging import debug_literal
@@ -66,12 +67,6 @@ def __init__(self, subparser):
             help="database lineages file",
         )
         p.add_argument("sketches", nargs="+", help="sketches to combine")
-        p.add_argument("--scaled", default=1000, type=int)
-        p.add_argument("-k", "--ksize", default=31, type=int)
-        p.add_argument(
-            "-m", "--moltype",
-            default="DNA"
-        )
         p.add_argument(
             "-o",
             "--output",
@@ -89,6 +84,7 @@ def __init__(self, subparser):
             action="store_true",
             help="Enable abundance tracking of hashes across rank selection.",
         )
+        sourmash_utils.add_standard_minhash_args(p)
 
     def main(self, args):
         super().main(args)
@@ -107,18 +103,13 @@ def __init__(self, subparser):
         p = subparser
 
         p.add_argument("sketches", nargs="+", help="sketches to combine")
-        p.add_argument("--scaled", default=1000, type=int)
-        p.add_argument("-k", "--ksize", default=31, type=int)
-        p.add_argument(
-            "-m", "--moltype",
-            default="DNA"
-        )
         p.add_argument(
             "-o",
             "--output",
             required=True,
             help="Define a filename for the pangenome signatures (.zip preferred).",
         )
+        sourmash_utils.add_standard_minhash_args(p)
 
     def main(self, args):
         super().main(args)
@@ -140,13 +131,6 @@ def __init__(self, subparser):
             metavar="SOURMASH_DATABASE",
             help="The sourmash dictionary created from 'pangenome_creatdb --abund'",
         )
-        p.add_argument(
-            "-k",
-            "--ksize",
-            type=int,
-            default=31,
-            help="The ksize of the sourmash pangenome database",
-        )
         p.add_argument(
             "-l",
             "--lineage",
@@ -164,6 +148,7 @@ def __init__(self, subparser):
             required=False,
             help="CSV file containing classification of each hash",
         )
+        sourmash_utils.add_standard_minhash_args(p)
 
     def main(self, args):
         super().main(args)
@@ -181,9 +166,9 @@ def __init__(self, subparser):
         super().__init__(subparser)
         p = subparser
         p.add_argument("metagenome_sig")
-        p.add_argument("-k", "--ksize", default=31, help="k-mer size", type=int)
         p.add_argument("ranktable_csv_files", nargs="+",
                        help="rank tables produced by pangenome_ranktable")
+        sourmash_utils.add_standard_minhash_args(p)
 
     def main(self, args):
         super().main(args)
@@ -207,14 +192,13 @@ def pangenome_createdb_main(args):
     if args.csv:
         csv_file = check_csv(args.csv)
 
+    select_mh = sourmash_utils.create_minhash_from_args(args)
+    print(f"selecting sketches: {select_mh}")
+
     # Load the database
     for filename in args.sketches:
-        print(f"loading file {filename} as index => manifest")
-        db = sourmash_args.load_file_as_index(filename)
-        db = db.select(ksize=args.ksize)
-        # @CTB check moltype
-        mf = db.manifest
-        assert mf, "no matching sketches for given ksize!?"
+        print(f"loading sketches from file {filename}")
+        db = sourmash_utils.load_index_and_select(filename, select_mh)
 
         if args.csv:
             chunk = []
@@ -322,7 +306,7 @@ def pangenome_createdb_main(args):
 
                 ss = sourmash.SourmashSignature(abund_mh, name=sig_name)
             else:
-                ss = sourmash/SourmashSignature(mh, name=sig_name)
+                ss = sourmash.SourmashSignature(mh, name=sig_name)
 
             save_sigs.add(ss)
 
@@ -352,14 +336,13 @@ def check_csv(csv_file):
 #
 
 def pangenome_merge_main(args):
+    select_mh = sourmash_utils.create_minhash_from_args(args)
+    print(f"selecting sketches: {select_mh}")
+
     # Load the database
     for filename in args.sketches:
-        print(f"loading file {filename} as index => manifest")
-        db = sourmash_args.load_file_as_index(filename)
-        db = db.select(ksize=args.ksize)
-        # @CTB check moltype
-        mf = db.manifest
-        assert mf, "no matching sketches for given ksize!?"
+        print(f"loading sketches from file {filename}")
+        db = sourmash_utils.load_index_and_select(filename, select_mh)
 
         c = Counter()
         mh = None
@@ -393,21 +376,18 @@ def pangenome_merge_main(args):
 
 
 def db_process(
-    filename,
-    ignore_case,
-    k=31,
-    lineage_name="None",
+        filename,
+        ignore_case,
+        select_mh=None,
+        lineage_name="None",
 ):
-    bname = os.path.basename(filename)
-    ss_dict = {}
-    print(f"\nloading file {bname} as index => manifest")
+    print(f"selecting sketches: {select_mh}")
 
-    db = sourmash_args.load_file_as_index(filename)
-    db = db.select(ksize=k)
-    mf = db.manifest
-    print(f"{bname} contains {len(mf)} signatures")
+    ss_dict = {}
+    print(f"loading sketches from file '{filename}'")
+    db = sourmash_utils.load_index_and_select(filename, select_mh)
+    print(f"'{filename}' contains {len(db)} signatures")
 
-    assert mf, "no matching sketches for given ksize!?"
     if lineage_name:
         print(f"Looking for {lineage_name} signature\n")
 
@@ -422,10 +402,11 @@ def search_pattern(vals):
             return any(pattern.search(val) for val in vals)
 
         # find all matching rows.
+        mf = db.manifest
         sub_mf = mf.filter_on_columns(search_pattern, ["name", "filename", "md5"])
 
         selected_sigs = []
-        print(f"Found {len(sub_mf)} signatures in {bname}:")
+        print(f"Found {len(sub_mf)} signatures in '{filename}':")
 
         for n, row in enumerate(sub_mf.rows, start=1):
             print(f'{n:<15} \033[0;31m{row.get("name")}\033[0m')
@@ -441,7 +422,7 @@ def search_pattern(vals):
         for ss in db.signatures():
             name = ss.name
 
-            print(f"Found \033[0;31m{name}\033[0m in {bname}")
+            print(f"Found \033[0;31m{name}\033[0m in '{filename}'")
 
             mh = ss.minhash
             hashes = mh.hashes
@@ -514,9 +495,11 @@ def pangenome_elements(data):
 #
 
 def pangenome_ranktable_main(args):
+    select_mh = sourmash_utils.create_minhash_from_args(args)
+
     ss_dict = db_process(
         filename=args.data,
-        k=args.ksize,
+        select_mh=select_mh,
         lineage_name=args.lineage,
         ignore_case=args.ignore_case,
     )
@@ -547,8 +530,10 @@ def pangenome_ranktable_main(args):
 #
 
 def classify_hashes_main(args):
-    db = sourmash.load_file_as_index(args.metagenome_sig)
-    db = db.select(ksize=args.ksize)
+    select_mh = sourmash_utils.create_minhash_from_args(args)
+    print(f"selecting sketches: {select_mh}")
+
+    db = sourmash_utils.load_index_and_select(args.metagenome_sig, select_mh)
     sketches = list(db.signatures())
     assert len(sketches) == 1
     sketch = sketches[0]

diff --git a/test_workflow/Makefile b/test_workflow/Makefile
@@ -1,6 +1,8 @@
+.PHONY: all clean cleanall
 all:
 	snakemake -c 4 -p
 
-cleanall:
+clean:
 	snakemake -c 4 --delete-all-output
-	snakemake -c 4 -p
+
+cleanall: clean all
diff --git a/test_workflow/Snakefile b/test_workflow/Snakefile
@@ -24,7 +24,7 @@ rule merge_agatha:
         pangenome_sig="test_output/agatha-merged.sig.zip",
     shell: """
         sourmash scripts pangenome_createdb {input.db} -t {input.tax} \
-            -o {output} --abund -k 21
+            -o {output} --abund -k 21 --dna --scaled=1000
     """
 
 rule merge_agatha_2: