Add pfam download

dbman/main.py

@@ -164,6 +164,7 @@ def run_in_tempdir(func=None, success=None, fail=None):
 
 def _recipes(args):
    recipes = {
+     'pfam': {'download': {'script': _pkgres('recipes/download_pfam.py')}},
      'card': {'download': {'script': _pkgres('recipes/download_card.py')},
               'blast': {'script': _pkgres('recipes/create_blast_dbs.py')},
               'diamond': {'script': _pkgres('recipes/create_diamond_dbs.py')},

dbman/recipes/download_pfam.py

+#!/usr/bin/env python3
+import dbman.helper as h
+import json
+import os
+import gzip
+
+import logging
+logging.basicConfig(level=logging.INFO)
+
+dbname = "pfam"
+dbdescription = "The Pfam database is a large collection of protein families"
+url = "ftp://ftp.ebi.ac.uk/pub/databases/Pfam/current_release"
+
+def read_version_info(version_fn):
+  version_data = {}
+  with gzip.open(version_fn, 'rt') as f:
+    for line in f:
+      (key, value) = line.split(':')
+      key = key.strip()
+      value = value.strip()
+      if key == "Pfam release":
+        key = "version"
+      if key == "Pfam-A families":
+        key = "families"
+      if key == "Date":
+        key = "release_date"
+      if key == "Based on UniProtKB":
+        key = "based_on_uniprotkb_version"
+      version_data[key] = value
+  return version_data
+
+version_fn = h.download(url + '/Pfam.version.gz')
+version_info = read_version_info(version_fn)
+
+# download archive
+hmm_fn = h.download(url + "/Pfam-A.hmm.gz")
+flatfile_fn = h.download(url + "/Pfam-A.hmm.dat.gz")
+
+# lookup tags for fasta files
+parts = []
+parts.append({'files':[hmm_fn], 'tags': ['profile', 'hmm', 'gzip', 'compressed']})
+parts.append({'files':[flatfile_fn], 'tags': ['flatfile', 'annotation', 'gzip', 'compressed']})
+parts.append({'files':[version_fn], 'tags': ['version_info', 'gzip', 'compressed']})
+version_info['parts'] = parts
+
+# write metadata file
+h.create_metadata(dbname, "download", dbdescription, other=version_info)