Add CI for wheel upload

.github/workflows/build_wheels.yml

@@ -0,0 +1,55 @@
+name: Build and upload to PyPI
+
+on:
+  release:
+    types:
+      - published
+
+jobs:
+  build_wheels:
+    name: Build wheels on ${{ matrix.os }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        # macos-13 is an intel runner, macos-14 is apple silicon
+        os: [ubuntu-latest, macos-13, macos-14]
+
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        submodules: 'recursive'
+
+    - name: Build wheels
+      uses: pypa/cibuildwheel@v2.22.0
+      env:
+        CIBW_BEFORE_ALL_MACOS: brew install autoconf automake
+        CIBW_BEFORE_ALL_LINUX: yum install -y bzip2-devel || apk add --upgrade bzip2-dev || apt-get install -y libbz2-dev
+        # Skip these architectures, 32-bit, PyPy, and musl
+        CIBW_SKIP: '*-musllinux_i686 *-manylinux_i686 pp* *musllinux*'
+
+    - uses: actions/upload-artifact@v4
+      with:
+        name: cibw-wheels-${{ matrix.os }}-${{ strategy.job-index }}
+        path: ./wheelhouse/*.whl
+
+  pypi-publish:
+    needs: [build_wheels]
+    name: Upload release to PyPI
+    runs-on: ubuntu-latest
+    environment:
+      name: pypi
+      url: https://pypi.org/p/<your-pypi-project-name>
+    permissions:
+      id-token: write
+    steps:
+    - uses: actions/download-artifact@v4
+      with:
+        pattern: cibw-*
+        path: dist
+        merge-multiple: true
+
+    - name: Publish package distributions to PyPI
+      uses: pypa/gh-action-pypi-publish@release/v1
+      # Testing only
+      # with:
+      #   repository-url: https://test.pypi.org/legacy/

EASTR/extract_junctions.py

@@ -1,11 +1,16 @@
 import collections
 import multiprocessing
 import os
+import pathlib
 import shlex
 import subprocess
 
 import pandas as pd
 
+this_directory = pathlib.Path(__file__).resolve().parent
+# This should exist with source after compilation.
+JUNCTION_CMD = os.path.join(this_directory, 'junction_extractor')
+
 
 def get_junctions_from_bed(bed_path: str) -> dict:
     junctions = {}
@@ -48,8 +53,9 @@ def get_junctions_multi_bed(bed_list:list, p) -> dict:
     return dd
 
 def junction_extractor(bam_path:str, out_path:str) -> dict:
+    check_for_dependency()
     name = os.path.splitext(os.path.basename(bam_path))[0]
-    cmd = f"junction_extractor -o {out_path} {bam_path}"
+    cmd = f"{JUNCTION_CMD} -o {out_path} {bam_path}"
     a = subprocess.Popen(shlex.split(cmd), stdout=subprocess.DEVNULL)
     b = a.communicate()
 
@@ -155,3 +161,7 @@ def extract_splice_sites_gtf(gtf_path:str) -> dict:
 #     x=extract_splice_sites_bam_regtools(bam_path)
 #     end = time.time()
 #     print(f"took {(end-start)/60} mins to get junctions")
+
+def check_for_dependency():
+    if not os.path.exists(JUNCTION_CMD):
+        raise RuntimeError(f"{JUNCTION_CMD} not found.")

EASTR/output_utils.py

@@ -2,6 +2,7 @@
 import csv
 import multiprocessing
 import os
+import pathlib
 import shlex
 import subprocess
 import sys
@@ -11,6 +12,10 @@
 from EASTR import alignment_utils
 from EASTR import utils
 
+this_directory = pathlib.Path(__file__).resolve().parent
+# This should exist with source after compilation.
+VACUUM_CMD = os.path.join(this_directory, 'vacuum')
+
 def out_junctions_filelist(bam_list:list, gtf_path, bed_list, out_junctions, suffix="") -> Union[List[str], None, str]:
     if out_junctions is None:
         return None
@@ -217,7 +222,8 @@ def spurious_dict_bam_by_sample_to_bed(spurious_dict, bam_list, out_removed_junc
 
 
 def filter_bam_with_vacuum(bam_path, spurious_junctions_bed, out_bam_path, verbose, removed_alignments_bam):
-    vacuum_cmd = "vacuum --remove_mate "
+    check_for_dependency()
+    vacuum_cmd = f"{VACUUM_CMD} --remove_mate "
     if verbose:
         vacuum_cmd = f"{vacuum_cmd} -V"
     if removed_alignments_bam:
@@ -289,3 +295,8 @@ def filter_multi_bam_with_vacuum(bam_list, sample_to_bed, out_bam_list, p, verbo
 #     # spurious_alignments, NH = get_spurious_alignments(bam_path, spurious_introns)
 #     # end = time.time()
 #     # print(f"took {(end-start)/60} mins"))
+
+def check_for_dependency():
+    """Check if runtime dependency exists."""
+    if not os.path.exists(VACUUM_CMD):
+        raise RuntimeError(f"{VACUUM_CMD} not found.")

README.md

@@ -27,64 +27,21 @@ regardless of the alignment software used.
 The installation steps for running EASTR are outlined below. Installation takes
 a few minutes.
 
-1. Clone source repository
+1. Clone repository
 
    ```bash
    git clone --recurse-submodules https://github.com/ishinder/EASTR.git
-   cd EASTR
    ```
 
-2. Compile junction_extractor and vacuum
+2. Install source code
 
-   ```bash
-   cd utils
-   cmake -DCMAKE_BUILD_TYPE=Release .
-   make
-   ```
-
-3. To add the junction_extractor and vacuum executables in the `utils` folder to
-   your PATH, you can follow these steps:
-
-   Get the absolute path of the build directory:
-
-   ```bash
-   pwd
-   ```
-
-   This command will print the absolute path of the current directory, which is
-   the `utils` directory.
-
-   Add the absolute path to the PATH environment variable: For temporary use
-   (not persistent across sessions), you can run:
-
-   ```bash
-   export PATH=$PATH:<absolute_path_to_utils_directory>
-   ```
-
-   Replace `<absolute_path_to_utils_directory>` with the path you obtained in
-   the previous step.
-
-   For persistent use (across sessions), you can add the `export` command to
-   your shell's configuration file. For bash, this is typically the `.bashrc` or
-   `.bash_profile` file in your home directory. For zsh, this is the `.zshrc`
-   file. To add the path to your shell configuration file, run:
-
-   ```bash
-   echo 'export PATH=$PATH:<absolute_path_to_utils_directory>' >> ~/.bashrc
-   ```
-
-4. Install EASTR
-
-   Ensure you are in the EASTR main directory before proceeding:
-
-   ```bash
-   # Navigate back to the main EASTR directory if you are not already there
-   cd path/to/EASTR
-   # (OPTIONAL) Install in a Python virtual environment
-   # python3 -m virtualenv .venv # (OPTIONAL)
-   # source .venv/bin/activate # (OPTIONAL)
-   make install # Install EASTR package
-   ```
+  ```bash
+  cd EASTR
+  # python3 -m venv .venv # (OPTIONAL)
+  # source .venv/bin/activate # (OPTIONAL)
+  pip install -U pip setuptools
+  pip install .
+  ```
 
 ### Required Arguments
 

setup.py

@@ -54,7 +54,7 @@ def build_extension(self, ext):
     subprocess.check_call(['cmake', '--build', '.', *build_args],
                           cwd=self.build_temp)
 
-description = 'Tool for emending alignments of spuriously spliced transcript reads'
+desc = 'Tool for emending alignments of spuriously spliced transcript reads'
 
 with open('./README.md', 'r', encoding='utf-8') as fh:
   long_description = fh.read()
@@ -67,7 +67,7 @@ def build_extension(self, ext):
     version='1.0.0',
     author='Ida Shinder',
     author_email='ishinde1@jhmi.edu',
-    description=description,
+    description=desc,
     long_description=long_description,
     long_description_content_type='text/markdown',
     url='https://github.com/ishinder/EASTR',
@@ -90,5 +90,6 @@ def build_extension(self, ext):
       ]
     },
     python_requires='>=3.10',
-    license = license_str,
+    license='MIT',
+    license_files=('LICENSE',),
 )