Merge pull request #11 from andrewjpage/master

Update ESKAPE databases with latest genomes from RefSeq
quadram-institute-bioscience · Jul 14, 2019 · c13cddb · c13cddb
2 parents 20ac159 + 2007df1
commit c13cddb
Show file tree

Hide file tree

Showing 36 changed files with 63 additions and 28 deletions.
diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-2.1.4
+2.1.6
diff --git a/scripts/socru b/scripts/socru
@@ -16,8 +16,10 @@ except pkg_resources.DistributionNotFound:
 	version = 'x.y.z'
 
 parser = argparse.ArgumentParser(
-	description = 'calculate the order and orientation of complete bacterial genomes',
-	usage = 'socru [options] species assembly.fasta',
+    description = '''Please cite our paper, "Socru: Typing of genome level order and orientation in bacteria", 
+    Andrew J Page, Gemma Langridge, 
+    bioRxiv 543702; (2019) doi: https://doi.org/10.1101/543702''',
+    usage = 'socru [options] species assembly.fasta',
     formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 
 parser.add_argument('species',  help='Species name, use socru_species to see all available', type=str)

diff --git a/scripts/socru_create b/scripts/socru_create
@@ -14,9 +14,12 @@ except pkg_resources.DistributionNotFound:
 	version = 'x.y.z'
 
 parser = argparse.ArgumentParser(
-	description = 'create genome arrangement type scheme',
-	usage = 'socru_create [options] output_directory assembly.fasta'
-    , formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    description = '''
+    Create genome arrangement type scheme.  Please cite our paper, "Socru: Typing of genome level order and orientation in bacteria", 
+    Andrew J Page, Gemma Langridge, 
+    bioRxiv 543702; (2019) doi: https://doi.org/10.1101/543702''',
+    usage = 'socru_create [options] output_directory assembly.fasta', 
+    formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 
 parser.add_argument('output_directory', help='Output directory', type=str)
 parser.add_argument('input_file', help='Input FASTA file (optionally gzipped)', type=str)

diff --git a/scripts/socru_lookup b/scripts/socru_lookup
@@ -16,9 +16,12 @@ except pkg_resources.DistributionNotFound:
 	version = 'x.y.z'
 
 parser = argparse.ArgumentParser(
-	description = 'Given a set of fragments, output the type',
-	usage = 'socru_lookup[options] /path/to/database 1-2-3-4-5-6-7'
-    , formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    description = '''
+    "Given a set of fragments, output the type. Please cite our paper, Socru: Typing of genome level order and orientation in bacteria", 
+    Andrew J Page, Gemma Langridge, 
+    bioRxiv 543702; (2019) doi: https://doi.org/10.1101/543702''',
+    usage = 'socru_lookup[options] /path/to/database 1-2-3-4-5-6-7', 
+    formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 
 parser.add_argument('db_dir',  help='Database directory', type=str )
 parser.add_argument('fragments', help="Fragments such as 1-2-3-4-5-6-7 or 1'-3-4'-5'-6'-7-2", type=str)

diff --git a/scripts/socru_rebuild_profile b/scripts/socru_rebuild_profile
@@ -14,8 +14,11 @@ except pkg_resources.DistributionNotFound:
 	version = 'x.y.z'
 
 parser = argparse.ArgumentParser(
-	description = 'Admin utility which will take in an existing profile and rebuild it',
-	usage = 'socru_rebuild_profile [options]',
+    description = '''
+    "Admin utility which will take in an existing profile and rebuild it. Please cite our paper, Socru: Typing of genome level order and orientation in bacteria", 
+    Andrew J Page, Gemma Langridge, 
+    bioRxiv 543702; (2019) doi: https://doi.org/10.1101/543702''',
+    usage = 'socru_rebuild_profile [options]',
     formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 
 parser.add_argument('profile_filename', help="profile.txt from database", type=str)
@@ -30,5 +33,3 @@ options = parser.parse_args()
 
 g = SocruRebuildProfile(options)
 g.run()
-
-
diff --git a/setup.py b/setup.py
@@ -18,7 +18,7 @@ def read(fname):
 	long_description=read('README.md'),
     packages = find_packages(),
     author='Andrew J. Page',
-    author_email='path-help@sanger.ac.uk',
+    author_email='andrew.page@quadram.ac.uk',
     url='https://github.com/quadram-institute-bioscience/socru',
     scripts=glob.glob('scripts/*'),
     test_suite='nose.collector',

diff --git a/socru/data/Acinetobacter_baumannii/profile.txt b/socru/data/Acinetobacter_baumannii/profile.txt
@@ -1,6 +1,7 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6
 1.0	1	2	3	4	5	6
 1.1	1'	2	3	4	5	6
+1.8	1	2	3	4'	5	6
 1.9	1'	2	3	4'	5	6
 2.24	1	2	3	5'	4'	6
 2.25	1'	2	3	5'	4'	6
diff --git a/socru/data/Enterobacter_asburiae/1.fa.gz b/socru/data/Enterobacter_asburiae/1.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/2.fa.gz b/socru/data/Enterobacter_asburiae/2.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/3.fa.gz b/socru/data/Enterobacter_asburiae/3.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/4.fa.gz b/socru/data/Enterobacter_asburiae/4.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/5.fa.gz b/socru/data/Enterobacter_asburiae/5.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/6.fa.gz b/socru/data/Enterobacter_asburiae/6.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/7.fa.gz b/socru/data/Enterobacter_asburiae/7.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/8.fa.gz b/socru/data/Enterobacter_asburiae/8.fa.gz
diff --git a/socru/data/Enterobacter_asburiae/profile.txt b/socru/data/Enterobacter_asburiae/profile.txt
@@ -1,2 +1,4 @@
-GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7
-1.0	1	2	3	4	5	6	7
+GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7	Frag_8
+1.0	1	2	3	4	5	6	7	8
+1.1	1'	2	3	4	5	6	7	8
+
diff --git a/socru/data/Enterobacter_asburiae/profile.txt.yml b/socru/data/Enterobacter_asburiae/profile.txt.yml
@@ -1,3 +1,3 @@
-dnaa_forward_orientation: true
-dnaa_fragment: 6
-reference_genome: Enterobacter/asburiae/LF7a/GCF_000224675.1_ASM22467v1_genomic.fna.gz
+dnaa_forward_orientation: false
+dnaa_fragment: 4
+reference_genome: Enterobacter/asburiae/asburiae_L1/GCF_000632395.1_ASM63239v1_genomic.fna.gz
diff --git a/socru/data/Enterobacter_cloacae/1.fa.gz b/socru/data/Enterobacter_cloacae/1.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/2.fa.gz b/socru/data/Enterobacter_cloacae/2.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/3.fa.gz b/socru/data/Enterobacter_cloacae/3.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/4.fa.gz b/socru/data/Enterobacter_cloacae/4.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/5.fa.gz b/socru/data/Enterobacter_cloacae/5.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/6.fa.gz b/socru/data/Enterobacter_cloacae/6.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/7.fa.gz b/socru/data/Enterobacter_cloacae/7.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/8.fa.gz b/socru/data/Enterobacter_cloacae/8.fa.gz
diff --git a/socru/data/Enterobacter_cloacae/profile.txt b/socru/data/Enterobacter_cloacae/profile.txt
@@ -1,3 +1,4 @@
-GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7	Frag_8
-1.0	1	2	3	4	5	6	7	8
-2.57	1'	2	3	6'	5'	4'	7	8
+GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7	Frag_8
+1.0	1	2	3	4	5	6	7	8
+1.1	1'	2	3	4	5	6	7	8
+2.57	1'	2	3	6'	5'	4'	7	8
diff --git a/socru/data/Enterobacter_cloacae/profile.txt.yml b/socru/data/Enterobacter_cloacae/profile.txt.yml
@@ -1,3 +1,3 @@
 dnaa_forward_orientation: true
 dnaa_fragment: 6
-reference_genome: Enterobacter/cloacae/ATCC_13047/GCF_000025565.1_ASM2556v1_genomic.fna.gz
+reference_genome: Enterobacter/cloacae/cloacae_ATCC_13047/GCF_000025565.1_ASM2556v1_genomic.fna.gz
diff --git a/socru/data/Enterobacter_hormaechei/profile.txt b/socru/data/Enterobacter_hormaechei/profile.txt
@@ -1,3 +1,4 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7	Frag_8
 1.0	1	2	3	4	5	6	7	8
-2.131	1'	8'	3	4	5	6	7	2'
+2.0	1	2	5	3	4	6	7	8
+3.131	1'	8'	3	4	5	6	7	2'
diff --git a/socru/data/Enterobacter_sp./profile.txt b/socru/data/Enterobacter_sp./profile.txt
@@ -1,2 +1,3 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7
 1.0	1	2	3	4	5	6	7
+2.24	1	2	3	5'	4'	6	7
diff --git a/socru/data/Enterococcus_faecium/profile.txt b/socru/data/Enterococcus_faecium/profile.txt
@@ -5,4 +5,5 @@ GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6
 2.0	6	1	2	3	4	5
 3.24	1	5'	4'	2	3	6
 4.32	1	6'	2	3	4	5
+4.33	1'	6'	2	3	4	5
 5.35	1'	6'	3	4	5	2'
diff --git a/socru/data/Escherichia_coli/profile.txt b/socru/data/Escherichia_coli/profile.txt
@@ -1,8 +1,14 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7
 1.0	1	2	3	4	5	6	7
+1.1	1'	2	3	4	5	6	7
 1.2	1	2'	3	4	5	6	7
+1.24	1	2	3	4'	5'	6	7
+1.25	1'	2	3	4'	5'	6	7
+1.64	1	2	3	4	5	6	7'
+1.9	1'	2	3	4'	5	6	7
 1.123	1'	7'	6'	5'	4'	3	2'
 2.67	1'	7'	3	4	5	6	2'
+2.90	1	7'	3	4'	5'	6	2'
 3.25	1'	4'	5'	2	3	6	7
 4.96	1	7'	6'	2	3	4	5
 4.97	1'	7'	6'	2	3	4	5
@@ -14,3 +20,5 @@ GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7
 10.65	1'	7'	2	3	4	5	6
 11.59	1'	4'	6'	5'	3	7	2'
 12.56	1	2	3	6'	5'	4'	7
+13.113	1'	7'	6'	5'	2	3	4
+14.64	1	2	7'	3	4	5	6
diff --git a/socru/data/Klebsiella_pneumoniae/profile.txt b/socru/data/Klebsiella_pneumoniae/profile.txt
@@ -6,3 +6,6 @@ GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7	Frag_8
 3.24	1	2	5'	4'	3	6	7	8
 4.131	1'	8'	3	4	5	6	7	2'
 5.0	1	2	3	5	4	6	7	8
+6.3	1'	3	4	5	6	7	8	2'
+7.32	1	2	3	4	5	7	8	6'
+8.35	1'	6'	3	4	5	6	7	2'
diff --git a/socru/data/Listeria_monocytogenes/profile.txt b/socru/data/Listeria_monocytogenes/profile.txt
@@ -1,5 +1,7 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6
 1.0	1	2	3	4	5	6
+1.16	1	2	3	4	5'	6
 2.56	1	2	3	6'	5'	4'
 3.33	1'	6'	2	3	4	5
 4.48	1	2	3	4	6'	5'
+5.35	1'	6'	3	4	5	2'
diff --git a/socru/data/Pseudomonas_aeruginosa/profile.txt b/socru/data/Pseudomonas_aeruginosa/profile.txt
@@ -1,7 +1,7 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4
 1.0	1	2	3	4
+1.1	1'	2	3	4
 1.9	1'	2	3	4'
-2.1	2	3	4	1'
-3.2	1	3	4	2'
-3.3	1'	3	4	2'
-3.11	1'	3	4'	2'
+2.2	1	3	4	2'
+2.3	1'	3	4	2'
+2.11	1'	3	4'	2'
diff --git a/socru/data/Salmonella_enterica/profile.txt b/socru/data/Salmonella_enterica/profile.txt
@@ -14,6 +14,7 @@ GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6	Frag_7
 8.66	1	7'	3	4	5	6	2'
 8.67	1'	7'	3	4	5	6	2'
 9.2	1	3	4	5	6	7	2'
+9.3	1'	3	4	5	6	7	2'
 10.74	1	7'	4'	3	5	6	2'
 11.0	1	2	3	5	6	4	7
 11.1	1'	2	3	5	6	4	7

diff --git a/socru/data/Staphylococcus_aureus/profile.txt b/socru/data/Staphylococcus_aureus/profile.txt
@@ -1,3 +1,8 @@
 GS	Frag_1	Frag_2	Frag_3	Frag_4	Frag_5	Frag_6
 1.0	1	2	3	4	5
 2.0	1	2	3	4	5	6
+2.1	1'	2	3	4	5	6
+3.16	1	2	5'	3	4
+4.17	1'	5'	2	3	4
+5.3	1'	3	4	5	2'
+6.3	1'	3	4	5	6	2'