MRG: update sourmash and snakemake dependencies

.github/workflows/test.yml

@@ -8,7 +8,7 @@ jobs:
 
     strategy:
       matrix:
-        python: [3.9]
+        python: ["3.10"]
 
     name: Python ${{ matrix.python }}
 

charcoal/Snakefile

@@ -4,7 +4,7 @@
 # --configfile must be specified on command line.
 #
 # CTB TODO: bring CLI improvements from grist over (@toplevel, etc.)
-import csv, sys, os, json
+import csv, sys, os, json, urllib
 from snakemake.workflow import srcdir
 from charcoal.utils import CSV_DictHelper
 
@@ -217,7 +217,7 @@ class Checkpoint_HitListPairs:
 ###
 
 wildcard_constraints:
-    size="\d+",
+    size="\\d+",
     g='[a-zA-Z0-9._-]+'                   # should be everything but /
 
 # default rule: build report index

charcoal/compare_taxonomy.py

@@ -11,7 +11,8 @@
 
 import sourmash
 from sourmash.lca.command_index import load_taxonomy_assignments
-from sourmash.lca import LCA_Database, LineagePair
+from sourmash.lca import LCA_Database
+from sourmash.tax.tax_utils import LineagePair
 
 from . import utils
 from .lineage_db import LineageDB
@@ -124,16 +125,9 @@ def get_genome_taxonomy(matches_filename, database_list,
                         genome_sig_filename, provided_lineage,
                         tax_assign, match_rank, min_f_ident, min_f_major):
     # load the matches from prefetch as a picklist
-    picklist = sourmash.picklist.SignaturePicklist('prefetch')
-    try:
-        picklist.load(matches_filename, picklist.column_name)
-    except ValueError:
-        with open(matches_filename, 'rt') as fp:
-            contents = fp.read()
-            if not len(contents): # empty is ok.
-                picklist = None
-            else:
-                raise
+    picklist = sourmash.picklist.SignaturePicklist('prefetch',
+                                                   pickfile=matches_filename)
+    picklist.load(allow_empty=True)
 
     # load all of the matches in the database, as found by prefetch;
     # select on them; and then aggregate into MultiIndex.

charcoal/conf/env-reporting.yml

@@ -4,13 +4,13 @@ channels:
     - bioconda
     - defaults
 dependencies:
-    - python>=3.8
+    - python>=3.10
     - papermill>=2.1.2,<3
     - notebook>=6,<7
     - matplotlib=3.5.2
     - plotly>=4.9.0,<5
     - ipykernel
-    - sourmash>=4.4.3,<5
+    - sourmash>=4.8.4,<5
     - pip
     - pip:
       - git+https://github.com/dib-lab/charcoal.git

charcoal/conf/env-sourmash.yml

@@ -4,11 +4,10 @@ channels:
     - bioconda
     - defaults
 dependencies:
-    - python>=3.7,<3.10
     - pytest>6,<7
     - pytest-dependency>=0.5.1
     - pyyaml>5.3,<6
-    - sourmash>=4.4.3,<5
+    - sourmash>=4.8.4,<5
     - pip
     - pip:
       - git+https://github.com/dib-lab/charcoal.git

charcoal/contigs_list_contaminants.py

@@ -83,8 +83,9 @@ def main(args):
     genome_sig = sourmash.load_one_signature(args.genome_sig)
 
     # load the matches from prefetch as a picklist
-    picklist = sourmash.picklist.SignaturePicklist('prefetch')
-    picklist.load(args.matches_csv, picklist.column_name)
+    picklist = sourmash.picklist.SignaturePicklist('prefetch',
+                                                   pickfile=args.matches_csv)
+    picklist.load()
 
     # load all of the matches in the database, as found by prefetch;
     # select on them; and then aggregate into MultiIndex.
@@ -130,7 +131,10 @@ def main(args):
         ident = get_ident(ss)
         lineage = tax_assign[ident]
 
-        lca_db.insert(ss, ident=ident)
+        try:
+            lca_db.insert(ss, ident=ident)
+        except ValueError:
+            continue
         lin_db.insert(ident, lineage)
 
     print(f'loaded {len(siglist)} signatures & created LCA Database')

charcoal/contigs_search_taxonomy.py

@@ -32,16 +32,9 @@ def main(args):
     genome_sig = sourmash.load_one_signature(args.genome_sig)
 
     # load the matches from prefetch as a picklist
-    picklist = sourmash.picklist.SignaturePicklist('prefetch')
-    try:
-        picklist.load(args.matches_csv, picklist.column_name)
-    except ValueError:
-        with open(args.matches_csv, 'rt') as fp:
-            contents = fp.read()
-            if not len(contents): # empty is ok.
-                picklist = None
-            else:
-                raise
+    picklist = sourmash.picklist.SignaturePicklist('prefetch',
+                                                   pickfile=args.matches_csv)
+    picklist.load(allow_empty=True)
 
     # load all of the matches in the database, as found by prefetch;
     # select on them; and then aggregate into MultiIndex.

charcoal/figs/sourmash_sankey.py

@@ -1,6 +1,7 @@
 from charcoal import utils
 import sourmash
-from sourmash.lca import taxlist, LineagePair
+from sourmash.lca import taxlist
+from sourmash.tax.tax_utils import LineagePair
 import collections
 
 import plotly.graph_objects as go

charcoal/just_taxonomy.py

@@ -13,7 +13,8 @@
 
 import sourmash
 from sourmash.lca.command_index import load_taxonomy_assignments
-from sourmash.lca import LCA_Database, LineagePair
+from sourmash.lca import LCA_Database
+from sourmash.tax.tax_utils import LineagePair
 
 from . import utils
 from . import lineage_db

charcoal/lineage_db.py

@@ -16,7 +16,7 @@
 import functools
 import pytest
 
-from sourmash.lca import LineagePair
+from sourmash.tax.tax_utils import LineagePair
 
 
 def cached_property(fun):
@@ -85,7 +85,7 @@ def __repr__(self):
     @classmethod
     def load(cls, db_name):
         "Load LCA_Database from a JSON file."
-        from .lca_utils import taxlist, LineagePair
+        from .lca_utils import taxlist
 
         xopen = open
         if db_name.endswith('.gz'):

charcoal/notebooks/report-genome.ipynb

@@ -39,7 +39,7 @@
     "\n",
     "import plotly.graph_objects as go\n",
     "\n",
-    "import sourmash\n",
+    "import sourmash, sourmash.lca.lca_utils\n",
     "from charcoal import utils\n",
     "from charcoal.figs.sourmash_sankey import GenomeSankeyFlow"
    ]
@@ -335,7 +335,7 @@
     "    if genome_lineage[-1].rank != 'genus':\n",
     "        genome_lineage = list(genome_lineage)\n",
     "        for rank in taxlist[len(genome_lineage):]:\n",
-    "            genome_lineage.append(sourmash.lca.LineagePair(rank=rank, name='unassigned'))\n",
+    "            genome_lineage.append(sourmash.lca.lca_utils.LineagePair(rank=rank, name='unassigned'))\n",
     "    \n",
     "    # set the color of the main lineage\n",
     "    genome_lineage = tuple(genome_lineage)\n",

charcoal/utils.py

@@ -6,7 +6,8 @@
 import csv
 
 import sourmash
-from sourmash.lca import lca_utils, LineagePair, taxlist, display_lineage
+from sourmash.lca import lca_utils, taxlist, display_lineage
+from sourmash.tax.tax_utils import LineagePair
 from sourmash.lca.lca_utils import (is_lineage_match, pop_to_rank,
                                     make_lineage)
 
@@ -246,8 +247,9 @@ def load_contamination_summary(fp):
     for k, items in x.items():
         z = []
         for source, target, count in items:
-            source = tuple([ LineagePair(rank, name) for rank, name in source ])
-            target = tuple([ LineagePair(rank, name) for rank, name in target ])
+            # get rank, name for each
+            source = tuple([ LineagePair(x[0], x[1]) for x in source ])
+            target = tuple([ LineagePair(x[0], x[1]) for x in target ])
             z.append((source, target, count))
         contam_d[k] = z
 

charcoal/version.py

@@ -0,0 +1 @@
+version = "0.1-dev"