Release 2.2.2 (#359)

* 🎨 Add snakefile

* 🔥 remove status-sacct.sh

* Fix Issue#350 (#353)

* Fix pandas version

* :up-arrow: Update pandas to 1.5
🎨 Chage line_terminator to lineterminator in `summary.py`

* Fix biopython error

* Update bioython version

* Fix ncbi_is_connected error (#352)

* Update metagenome.py

* Update metagenome.py

* Update version

* Remove snakemake workflow

* Remove space

Author: Sidduppal

VERSION

@@ -1 +1 @@
-2.2.1
+2.2.2

autometa-env.yml

@@ -6,14 +6,14 @@ channels:
 dependencies:
   - attrs # test-data requirement
   - bedtools
-  - biopython
+  - biopython>=1.82
   - bowtie2
   - diamond>=2.0
   - gdown
   - hmmer
   - numba>=0.47
   - numpy>=1.13
-  - pandas>=1.1
+  - pandas>=1.5
   - parallel
   - pip
   - prodigal # NOTE: 2.5 and 2.6 output format is different for sequence headers

autometa/binning/summary.py

@@ -256,7 +256,7 @@ def get_metabin_taxonomies(
     outcols = [cluster_col, "length", "taxid", *canonical_ranks]
     tmp_lines = (
         bin_df[outcols]
-        .to_csv(sep="\t", index=False, header=False, line_terminator="\n")
+        .to_csv(sep="\t", index=False, header=False, lineterminator="\n")
         .split("\n")
     )
     taxonomies = {}

autometa/common/metagenome.py

@@ -139,7 +139,7 @@ def length_weighted_gc(self) -> float:
         for seq in self.sequences:
             weight = len(seq) / self.size
             add_weight(weight)
-            gc_content = SeqUtils.GC(seq)
+            gc_content = SeqUtils.gc_fraction(seq) * 100
             add_gc_contents(gc_content)
         return np.average(a=gc_contents, weights=weights)
 
@@ -312,7 +312,7 @@ def gc_content(self) -> pd.DataFrame:
             [
                 {
                     "contig": record.id,
-                    "gc_content": SeqUtils.GC(record.seq),
+                    "gc_content": SeqUtils.gc_fraction(record.seq) * 100,
                     "length": len(record.seq),
                 }
                 for record in self.seqrecords

autometa/config/databases.py

@@ -200,9 +200,10 @@ def get_remote_checksum(self, section: str, option: str) -> str:
             raise ValueError(
                 f"'section' must be 'ncbi' or 'markers'. Provided: {section}"
             )
-        if not ncbi_is_connected():
-            raise ConnectionError("Cannot connect to the NCBI rsync server")
+        print(f"The section is : {section}")
         if section == "ncbi":
+            if not ncbi_is_connected():
+                raise ConnectionError("Cannot connect to the NCBI rsync server")
             host = self.config.get(section, "host")
             ftp_fullpath = self.config.get("checksums", option)
             chksum_fpath = ftp_fullpath.split(host)[-1]
@@ -604,7 +605,11 @@ def compare_checksums(self, section: str = None) -> Dict[str, Dict]:
                     # Skip user added options not required by Autometa
                     continue
                 # nodes.dmp, names.dmp and merged.dmp are all in taxdump.tar.gz
-                option = "taxdump" if option in {"nodes", "names", "merged", "delnodes"} else option
+                option = (
+                    "taxdump"
+                    if option in {"nodes", "names", "merged", "delnodes"}
+                    else option
+                )
                 fpath = self.config.get(section, option)
                 fpath_md5 = f"{fpath}.md5"
                 # We can not checksum a file that does not exist.

nextflow.config

@@ -12,7 +12,7 @@ manifest {
     doi = "https://doi.org/10.1093/nar/gkz148"
     mainScript = "main.nf"
     nextflowVersion = ">=21.04.0"
-    version = "2.2.1"
+    version = "2.2.2"
 }
 
 

tests/unit_tests/test_metagenome.py

@@ -69,7 +69,7 @@ def test_length_weighted_gc(metagenome):
     if metagenome.nseqs == 3:
         expected = 65.3167472932504
     elif metagenome.nseqs == 4:
-        expected = 63.195548489666145
+        expected = 63.19554848966614
     else:
         expected = None
     assert isinstance(metagenome.length_weighted_gc, float)