CI and dependencies managment (#20)

* fix: list looseversion in dependencies * fix: unpin biopython * fix: raise python pin to 3.13 * fix: GC roudings for tests with GC/gc_fraction * ci: add codecov back * docs: update badges
koszullab · Feb 20, 2025 · 5fd32e3 · 5fd32e3
1 parent 8b5437e
commit 5fd32e3
Show file tree

Hide file tree

Showing 6 changed files with 25 additions and 13 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -48,4 +48,5 @@ jobs:
         run: |
           ruff check . --select=E9,F63,F7,F82
           pytest --cov --cov-report=xml
+          codecov
         shell: micromamba-shell {0}
diff --git a/README.md b/README.md
@@ -2,9 +2,9 @@
 
 [![PyPI version](https://badge.fury.io/py/metator.svg)](https://badge.fury.io/py/metator)
 ![PyPI - Python Version](https://img.shields.io/pypi/pyversions/metator.svg)
-[![Build Status](https://github.com/koszullab/metator/actions/workflows/python-package.yml/badge.svg)](https://github.com/koszullab/metaTOR/actions)
+[![Build Status](https://github.com/koszullab/metator/actions/workflows/ci.yml/badge.svg)](https://github.com/koszullab/metaTOR/actions)
 [![codecov](https://codecov.io/gh/koszullab/metator/branch/master/graph/badge.svg)](https://codecov.io/gh/koszullab/metator)
-[![Docker Cloud Build Status](https://img.shields.io/docker/cloud/build/koszullab/metator)](https://hub.docker.com/r/koszullab/metator)
+<!-- [![Docker Cloud Build Status](https://img.shields.io/docker/cloud/build/koszullab/metator)](https://hub.docker.com/r/koszullab/metator) -->
 [![Read the docs](https://readthedocs.org/projects/metator/badge)](https://metator.readthedocs.io)
 [![License: GPLv3](https://img.shields.io/badge/License-GPL%203-0298c3.svg)](https://opensource.org/licenses/bo-3.0)
 [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/ambv/black)

diff --git a/metator.yaml b/metator.yaml
@@ -4,11 +4,11 @@ channels:
   - bioconda
   - defaults
 dependencies:
-  - python<3.12
+  - python<3.13
   - hicstuff
   - networkx
   - checkv
-  - biopython<=1.80
+  - biopython
   - pysam
   - pairtools
   - pyfastx
@@ -18,6 +18,7 @@ dependencies:
   - scikit-learn
   - scipy
   - seaborn
+  - looseversion
   - bowtie2
   - pairix
   - bwa

diff --git a/pyproject.toml b/pyproject.toml
@@ -7,7 +7,7 @@ name = "metator"
 version = "1.3.4"
 description = "A pipeline for binning metagenomic datasets from metaHiC data."
 readme = "README.md"
-requires-python = ">=3.9,<3.12"
+requires-python = ">=3.9,<3.13"
 license = { text = "GNU General Public License v3 (GPLv3)" }
 
 authors = [
@@ -34,7 +34,7 @@ dependencies = [
     "hicstuff",
     "networkx",
     "checkv",
-    "biopython<=1.80",
+    "biopython",
     "pysam",
     "pairtools",
     "pyfastx",
@@ -44,18 +44,17 @@ dependencies = [
     "scikit-learn",
     "scipy",
     "seaborn",
+    "looseversion",
     "micomplete"
 
     # NON PIP DEPENDENCIES
-
     #"bowtie2"
     #"pairix"
     #"bwa"
     #"samtools"
     #"prodigal"
     #"hmmer"
-    #"gcc"
-    #"java-jdk"
+    #"openjdk"
 ]
 
 [project.optional-dependencies]
@@ -90,6 +89,7 @@ docs = [
 dev = [
     "metator[test,docs]",
     "hatch",
+    "hatchling", 
     "hatch-build-scripts", 
     "pre-commit",
     "mypy",

diff --git a/src/metator/network.py b/src/metator/network.py
@@ -19,6 +19,8 @@
 import csv
 import numpy as np
 import re
+from packaging import version
+import Bio
 from Bio import SeqIO
 from Bio import SeqUtils
 from os.path import join, basename
@@ -294,10 +296,14 @@ def create_contig_data(assembly, nb_alignment=1, depth_file=None, enzyme=None):
             line = depth.readline()
             for contig in SeqIO.parse(assembly, "fasta"):
                 line = depth.readline().split("\t")
+                if version.parse(Bio.__version__) >= version.parse("1.80"):
+                    gc_content = int(SeqUtils.gc_fraction(contig.seq) * 100 * 10**13) / 10**13
+                else:
+                    gc_content = int(SeqUtils.GC(contig.seq) * 10**13) / 10**13
                 contig_data[contig.id] = {
                     "id": global_id,
                     "length": int(line[1]),
-                    "GC": SeqUtils.gc_fraction(contig.seq)*100,
+                    "GC": gc_content,
                     "hit": 0,
                     "coverage": float(line[2]),
                     "RS": (len(re.findall(pattern, str(contig.seq))) + 1)
@@ -312,10 +318,14 @@ def create_contig_data(assembly, nb_alignment=1, depth_file=None, enzyme=None):
                 global_id += 1
     else:
         for contig in SeqIO.parse(assembly, "fasta"):
+            if version.parse(Bio.__version__) >= version.parse("1.80"):
+                gc_content = int(SeqUtils.gc_fraction(contig.seq) * 100 * 10**13) / 10**13
+            else:
+                gc_content = int(SeqUtils.GC(contig.seq) * 10**13) / 10**13
             contig_data[contig.id] = {
                 "id": global_id,
                 "length": len(contig.seq),
-                "GC": SeqUtils.gc_fraction(contig.seq)*100,
+                "GC": gc_content,
                 "hit": 0,
                 "coverage": "-",
                 "RS": (len(re.findall(pattern, str(contig.seq))) + 1)

diff --git a/tests/test_network.py b/tests/test_network.py
@@ -128,7 +128,7 @@ def test_create_contig_data():
     assert contig_data["NODE_522"] == {
         "id": 1,
         "length": 22786,
-        "GC": 62.07759150355482,
+        "GC": 62.0775915035548,
         "hit": 0,
         "coverage": "-",
         "RS": "-",
@@ -142,7 +142,7 @@ def test_create_contig_data():
     assert contig_data["NODE_522"] == {
         "id": 1,
         "length": 22786,
-        "GC": 62.07759150355482,
+        "GC": 62.0775915035548,
         "hit": 0,
         "coverage": 4.76595,
         "RS": 162,