add assembly_summary that triggered the problems

Snakefile

@@ -3,22 +3,39 @@ from pathlib import Path
 SBT_ZIP = Path("/home/luizirber/work/sourmash-bio/sourmash_resources_debug/genbank-bacteria-x1e6-k51.sbt.zip")
 
 rule all:
-  input: "outputs/newtree.sbt.zip"
+  input:
+    "outputs/signames_new_catalog.txt",
+    "outputs/signames_new.txt"
 
-rule extract_sig_names:
-  output: "outputs/signames.txt"
-  input: SBT_ZIP
-  run:
+def extract(inp, out):
     import sourmash
 
-    index = sourmash.load_file_as_index(input[0])
-    with open(output[0], 'w') as out:
+    index = sourmash.load_file_as_index(inp)
+    with open(out, 'w') as out:
       for leaf in index.leaves():
         sig = leaf.data
         out.write(sig.name() + "\n")
         leaf.unload()
 
-rule build_sbt:
+rule extract_sig_names:
+  output: "outputs/signames.txt"
+  input: SBT_ZIP
+  run:
+    extract(input[0], output[0])
+
+rule extract_sig_names_new_sbt:
+  output: "outputs/signames_new.txt"
+  input: "outputs/newtree.sbt.zip"
+  run:
+    extract(input[0], output[0])
+
+rule extract_sig_names_new_sbt_catalog:
+  output: "outputs/signames_new_catalog.txt"
+  input: "outputs/newtree_catalog.sbt.zip"
+  run:
+    extract(input[0], output[0])
+
+rule build_sbt_from_set:
   output: "outputs/newtree.sbt.zip"
   input: "outputs/signames.txt"
   run:
@@ -38,3 +55,43 @@ rule build_sbt:
         print(f"processed {i} sigs")
 
     index.save(output[0])
+
+rule build_sbt_from_catalog:
+  output: "outputs/newtree_catalog.sbt.zip"
+  input: "outputs/catalog.txt"
+  run:
+    import sourmash
+
+    index = sourmash.create_sbt_index(bloom_filter_size=64)
+    signames = []
+    with open(input[0], 'r') as f:
+      for line in f:
+        signames.append(line.strip())
+
+    for i, signame in enumerate(signames):
+      mh = sourmash.MinHash(0, 31, scaled=2000, mins=[i])
+      new_sig = sourmash.SourmashSignature(mh, name=signame)
+      index.insert(new_sig)
+      if i % 1000 == 0:
+        print(f"processed {i} sigs")
+
+    index.save(output[0])
+
+rule generate_catalog:
+  output: "outputs/catalog.txt"
+  input: "outputs/assembly_summary_genbank.txt.gz"
+  run:
+    import csv
+    import gzip
+
+    with open(output[0], 'w') as out:
+      with gzip.open(input[0], 'rt') as fp:
+        fp.readline() # skip first line
+        fp.read(2) # skip initial comment in header
+        data = csv.DictReader(fp, delimiter='\t')
+        for row in data:
+          name_parts = [row["assembly_accession"], " ", row['organism_name']]
+          if row['infraspecific_name']:
+            name_parts += [" ", row['infraspecific_name']]
+          name_parts += [', ', row['asm_name']]
+          out.write("".join(name_parts) + "\n")