Prepare for bioconda (#21)

* fix: entirely remove `pairix` (not `pypairix`) dependency

* ci: move micomplete to conda deps

* ci: rm docker files

* ci: move artifacts to /src

* ci: comment out definition of $CXX variable

* bump to 1.3.6

.codecov.yml

@@ -4,6 +4,6 @@ coverage:
       default:
         target: auto
         threshold: 1%
-        if_ci_failed: ignore #success, failure, error, ignore
+        if_ci_failed: error #success, failure, error, ignore
         informational: false
         only_pulls: false

.gitignore

@@ -139,4 +139,5 @@ artifacts/
 gen-louvain/
 networkanalysis/
 pairix/
-bowtie2/
+bowtie2/
+bin/

CHANGELOG.md

@@ -1,8 +1,13 @@
-
 # Change Log
 
 All notable changes to this project will be documented in this file.
 
+## [1.3.6] - 2025-02-21
+- Prepare for bioconda release.
+- Only rely on python `pypairix` package (installable from pip). 
+- Binaries are now put into `./bin` instead of `./external/artifacts/`. 
+- micomplete is now avaible from `bioconda`. 
+
 ## [1.3.4] - 2025-02-19
 - Package now relies on `pyproject.toml` for build configuration with `hatch`. 
 - Binaries for `louvain` and `leiden` clustering algorithms are now embedded in the package.

MANIFEST.in

@@ -1,4 +1,4 @@
 include README.md
-include external/artifacts/*
+include bin/*
 recursive-exclude test_data *
 recursive-exclude tests *

README.md

@@ -4,7 +4,6 @@
 ![PyPI - Python Version](https://img.shields.io/pypi/pyversions/metator.svg)
 [![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) -->
 [![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)
@@ -44,7 +43,6 @@ before installing `metator`:
 * Python `3.9` to `3.11` is required.
 * The following dependencies should also be locally installed and available in the `$PATH`:
     * [`bowtie2`](http://bowtie-bio.sourceforge.net/bowtie2/index.shtml) or `bwa`
-    * [`pairix`](https://github.com/4dn-dcic/pairix)
     * [`samtools`](https://www.htslib.org/download/)
     * [`hmmer`](http://hmmer.org/documentation.html)
     * [`prodigal`](https://github.com/hyattpd/Prodigal)
@@ -53,42 +51,23 @@ before installing `metator`:
 * The following non-pythonic librairies are **embedded** when installing `metator` with `pip`: [`louvain 0.3`](https://sourceforge.net/projects/louvain/files/GenericLouvain/) and [`leiden 1.3.0`](https://github.com/CWTSLeiden/networkanalysis).
 
 ```sh
-# Install bowtie2, sameools, hmmer, prodigal and java-jdk:
+# Install bowtie2, samtools, hmmer, prodigal and java-jdk:
 sudo apt update && sudo apt install bowtie2 samtools hmmer prodigal default-jdk
 
-# Also install pairix:
-wget https://github.com/4dn-dcic/pairix/archive/refs/tags/0.3.9.zip -O pairix-0.3.9.zip
-unzip pairix-0.3.9.zip
-mv pairix-0.3.9 ~/.local/lib/pairix
-cd ~/.local/lib/pairix
-make
-chmod +x bin/pairix
-echo 'export PATH=$PATH:~/.local/lib/pairix/bin' >> ~/.bashrc
-
 # Install metator from Pypi
 pip3 install metator
 ```
 
 To use the development version:
 
 ```sh
-# Install bowtie2, sameools, hmmer, prodigal, java-jdk and pairix, see above
+# Install bowtie2, sameools, hmmer, prodigal, java-jdk, see above
 
 git clone https://github.com/koszullab/metator
 cd metator
 pip3 install -e .[dev]
 ```
 
-<!-- 
-### Using docker container
-
-A dockerfile is also available if that is of interest. You may fetch the image by running the following:
-
-```sh
-docker pull koszullab/metator
-``` 
--->
-
 ## Usage
 
 ```sh

external/setup_dependencies.sh

@@ -1,11 +1,12 @@
 #!/bin/bash
 
 ## Purge existing artifacts (required for local rebuild)
-rm -rf artifacts/ gen-louvain/ pairix/ bowtie2/ networkanalysis/
+rm -rf ../bin/ gen-louvain/ networkanalysis/
 
 ## Install louvain
 tar -k -xzf louvain-generic.tar.gz
 cd gen-louvain
+sed -i 's/^CXX=g++/#&/' Makefile
 make
 cd ..
 
@@ -14,30 +15,12 @@ cd ..
 mkdir -p networkanalysis/build/libs/
 cp networkanalysis-1.3.0.jar networkanalysis/build/libs/
 
-## Install pairix
-# wget https://github.com/4dn-dcic/pairix/archive/refs/tags/0.3.9.zip -O pairix-0.3.9.zip
-# zip -d pairix-0.3.9.zip "pairix-0.3.9/samples/*"
-# unzip pairix-0.3.9.zip
-# mv pairix-0.3.9 pairix
-# cd pairix
-# make
-# chmod +x bin/pairix
-# cd ..
-
-# ## Install bowtie2
-# wget https://sourceforge.net/projects/bowtie-bio/files/bowtie2/2.5.1/bowtie2-2.5.1-source.zip/download -O bowtie2-2.5.1-source.zip
-# zip -d bowtie2-2.5.1-source.zip "bowtie2-2.5.1/example/*"
-# unzip bowtie2-2.5.1-source.zip
-# mv bowtie2-2.5.1/ bowtie2
-# cd bowtie2
-# make
-# cd ..
-
 ## Move artifacts to the correct location
-mkdir -p artifacts/networkanalysis/build artifacts/pairix artifacts/bowtie2/bin
-mv gen-louvain/ artifacts/
-mv networkanalysis/build artifacts/networkanalysis/
-# mv pairix/* artifacts/pairix/
-# mv bowtie2/bowtie2* artifacts/bowtie2/bin/
+mkdir -p ../bin/
+mv gen-louvain/louvain ../bin/
+mv gen-louvain/convert ../bin/
+mv gen-louvain/hierarchy ../bin/
+mv gen-louvain/matrix ../bin/
+mv networkanalysis/build/libs/networkanalysis-1.3.0.jar ../bin/networkanalysis-1.3.0.jar
 
-rm -rf gen-louvain/ pairix/ bowtie2/ networkanalysis/
+rm -rf gen-louvain/ networkanalysis/

metator.yaml

@@ -27,6 +27,6 @@ dependencies:
   - hmmer
   - gcc
   - java-jdk
+  - micomplete
   - pip:
-    - micomplete
     - metator

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "metator"
-version = "1.3.4"
+version = "1.3.6"
 description = "A pipeline for binning metagenomic datasets from metaHiC data."
 readme = "README.md"
 requires-python = ">=3.9,<3.13"
@@ -45,11 +45,11 @@ dependencies = [
     "scipy",
     "seaborn",
     "looseversion",
-    "micomplete"
+    "micomplete",
+    "pypairix"
 
     # NON PIP DEPENDENCIES
     #"bowtie2"
-    #"pairix"
     #"bwa"
     #"samtools"
     #"prodigal"
@@ -114,13 +114,13 @@ path = "src/metator/version.py"
 allow-direct-references = true
 
 [[tool.hatch.build.hooks.build-scripts.scripts]]
-out_dir = "external/artifacts/"
+out_dir = "bin/"
 work_dir = "external"
 commands = ["bash setup_dependencies.sh"]
 artifacts = []
 
 [tool.hatch.build.force-include]
-"external/artifacts" = "metator/external/artifacts"
+"bin" = "metator/bin"
 
 [tool.black]
 line-length = 130

src/metator/__init__.py

@@ -6,9 +6,8 @@
 from .version import __version__ as version
 from . import *
 
-__author__ = "Amaury Bignaud, Jacques Serizay, Lyam Baudry, Théo Foutel-Rodier,\
-    Martial Marbouty"
-__copyright__ = "Copyright © 2017-2018, Institut Pasteur, Paris, France"
+__author__ = "Amaury Bignaud, Jacques Serizay, Lyam Baudry, Théo Foutel-Rodier, Martial Marbouty"
+__copyright__ = "Copyright © 2017-2025, Institut Pasteur, Paris, France"
 __credits__ = [
     "Amaury Bignaud",
     "Jacques Serizay",
@@ -40,6 +39,6 @@ def is_editable_install():
 __metator_root__ = __metator_source__
 if is_editable_install():
     __metator_root__ = os.path.abspath(os.path.join(__metator_source__, "../../"))
-__leiden_dir__ = Path(__metator_root__, "external", "artifacts", "networkanalysis", "build", "libs")
-LEIDEN_PATH = str(next(__leiden_dir__.glob("networkanalysis-1.3.0*.jar")))
-LOUVAIN_PATH = str(Path(__metator_root__, "external", "artifacts", "gen-louvain"))
+__bin_dir__ = Path(__metator_root__, "bin")
+LEIDEN_PATH = str(next(__bin_dir__.glob("networkanalysis-1.3.0*.jar")))
+LOUVAIN_PATH = str(__bin_dir__)

src/metator/commands.py

@@ -1372,7 +1372,7 @@ class Pairs(AbstractCommand):
     """Sort, compress and index pairs files for faster assess to the data.
 
     Sort the pairs file using pairtools. Compress them using bgzip. Index them
-    using pairix.
+    using pypairix.
 
     usage:
         pairs [--force] [--remove] [--threads=1] <pairsfile>...

src/metator/contact_map.py

@@ -237,9 +237,9 @@ def extract_pairs(metator_data):
                 )
             )
         for pairs_file in metator_data.pairs_files:
-            # Check if the pairix index exist.
+            # Check if the pypairix index exist.
             pairs_data = mio.get_pairs_data(pairs_file)
-            # Need a sorted (chr1 chr2 pos1 pos2) pair file indexed with pairix.
+            # Need a sorted (chr1 chr2 pos1 pos2) pair file indexed with pypairix.
             for contig_id1, contig in enumerate(metator_data.contigs):
                 # Only need to retrieve the upper triangle.
                 for contig_id2 in range(contig_id1, len(metator_data.contigs)):

src/metator/io.py

@@ -8,7 +8,7 @@
     - check_fasta_index
     - check_is_fasta
     - check_louvain_cpp
-    - check_pairix
+    - check_pypairix
     - check_pairtools
     - generate_fasta_index
     - generate_temp_dir
@@ -48,6 +48,7 @@
 from metator.log import logger
 from os.path import join, exists, isfile
 from random import getrandbits
+from packaging import version
 from packaging.version import Version
 
 
@@ -177,21 +178,20 @@ def check_louvain_cpp(louvain_path):
     return True
 
 
-def check_pairix():
+def check_pypairix():
     """
-    Function to test if pairix is in the path.
+    Function to test if pypairix is available.
 
     Returns:
     --------
     bool:
-        True if pairix found in the path, False otherwise.
+        True if pypairix is available.
     """
     try:
-        pairix = sp.check_output(f"pairix --help", stderr=sp.STDOUT, shell=True)
-    except sp.CalledProcessError:
-        logger.error("Cannot find 'pairix' in your path please install it or add it in your path.")
-        raise ImportError
-        return False
+        v = version.parse(pypairix.__version__)
+    except AttributeError:
+        logger.error("Cannot find 'pypairix' installed.")
+        raise AttributeError
     return True
 
 
@@ -293,15 +293,15 @@ def get_pairs_data(pairfile, threads=1, remove=False, force=False):
     str :
         Path to the sorted and indexed pair file.
     """
-    # Check if pairix index exists, generate it otherwise.
+    # Check if pypairix index exists, generate it otherwise.
     try:
         pairs_data = pypairix.open(pairfile)
     except pypairix.PairixError:
         try:
             pairfile_sorted = f"{os.path.splitext(pairfile)[0]}_sorted.pairs.gz"
             pairs_data = pypairix.open(pairfile_sorted)
         except pypairix.PairixError:
-            logger.warning("No pairix index found. Build the index.")
+            logger.warning("No pypairix index found. Build the index.")
             pairfile = sort_pairs_pairtools(pairfile, threads, remove, force)
             pairs_data = pypairix.open(pairfile)
     return pairs_data
@@ -771,8 +771,8 @@ def sort_pairs_pairtools(pairfile, threads=1, remove=False, force=False):
     # Extract basename of the file.
     basename = os.path.splitext(pairfile)[0]
 
-    # Test if pairix and pairtools are installed and in the path.
-    _ = check_pairix()
+    # Test if pypairix and pairtools are installed and in the path.
+    _ = check_pypairix()
     _ = check_pairtools()
 
     # Set the force parameter and delete files or raise an error accodringly.
@@ -800,9 +800,8 @@ def sort_pairs_pairtools(pairfile, threads=1, remove=False, force=False):
     process = sp.Popen(cmd, shell=True)
     _out, _err = process.communicate()
     # Indexed pairs.
-    cmd = f"set -eu ; pairix{force} {basename}_sorted.pairs.gz"
-    process = sp.Popen(cmd, shell=True)
-    _out, _err = process.communicate()
+    force_pypairix = 1 if force else 0
+    pypairix.build_index(f"{basename}_sorted.pairs.gz", force=force_pypairix)
 
     # Remove original pairfile if remove setup.
     if remove:

src/metator/main.py

@@ -26,7 +26,7 @@
                 final ouptut of metaTOR.
     scaffold    Scaffold a metator bin based on pairs files.
     pairs       Sort the pairs file using pairtools. Compress them using bgzip.
-                Index them using pairix.
+                Index them using pypairix.
     host        Detect bacterial host from a metaHiC network binned by metaTOR
                 given a annotated MGE list.
     mge         Build MGE MAGs based on metagenomic binning using metabat2

src/metator/mge.py

@@ -68,14 +68,14 @@ def build_matrix(
     mat = np.zeros((n, n))
     # Write one pair file for all the ones given.
     for pairs_file in pairs_files:
-        # Check if the pairix index exist
+        # Check if the pypairix index exist
         try:
             pairs_data = pypairix.open(pairs_file)
             pypairix_index = True
         except pypairix.PairixError:
-            logger.warning("No pairix index found. Iterates on the pairs.")
+            logger.warning("No pypairix index found. Iterates on the pairs.")
             pypairix_index = False
-        # Need a sorted (chr1 chr2 pos1 pos2) pair file indexed with pairix.
+        # Need a sorted (chr1 chr2 pos1 pos2) pair file indexed with pypairix.
         if pypairix_index:
             for i, contig in enumerate(contigs):
                 # Only need to retrieve the upper triangle.

tests/__init__.py

@@ -18,6 +18,6 @@ def is_editable_install():
 __metator_root__ = __metator_source__
 if is_editable_install():
     __metator_root__ = os.path.abspath(os.path.join(__metator_source__, "../../"))
-__leiden_dir__ = Path(__metator_root__, "external", "artifacts", "networkanalysis", "build", "libs")
-LEIDEN_PATH = str(next(__leiden_dir__.glob("networkanalysis-1.3.0*.jar")))
-LOUVAIN_PATH = str(Path(__metator_root__, "external", "artifacts", "gen-louvain"))
+__bin_dir__ = Path(__metator_root__, "bin")
+LEIDEN_PATH = str(next(__bin_dir__.glob("networkanalysis-1.3.0*.jar")))
+LOUVAIN_PATH = str(__bin_dir__)