Merge pull request #44 from RasmussenLab/dev

Dev

Author: joacjo

README.md

@@ -22,9 +22,19 @@ mamba install -c conda-forge -c bioconda snakemake
 mamba install -c conda-forge -c bioconda scikit-learn=1.0.2
 mamba install -c conda-forge -c bioconda cython
 mamba install -c conda-forge -c bioconda pygraphviz
+```
+
+
+```
+### Clone repository
+git clone the repository https://github.com/RasmussenLab/phamb.git
+
+### Quick install 
+pip install -e .
 
-### Old dependencies
-conda create -c conda-forge -c bioconda -n phamb python=3.6 cython scikit-learn=0.21.3 snakemake pygraphviz
+### Test installation
+mkdir -p testout 
+run_RF.py test/contigs.fna.gz test/clusters.tsv test testout
 ```
 
 
@@ -51,7 +61,7 @@ mkdir -p projectdir
 cd projectdir 
 git clone the repository https://github.com/RasmussenLab/phamb.git
 cp -r phamb/workflows/mag_annotation .
-python mag_annotation/scripts/split_contigs.py -c contigs.fna.gz 
+python split_contigs.py -c contigs.fna.gz 
 ```
 
 - Now the `contigs.fna.gz` is splitted into individual assemblies i.e. `assembly/{sample}/{sample}.fna`
@@ -98,7 +108,7 @@ gzip contigs.fna
 ### Run the RF model
 Running the provided script, the virome bins are written to a fasta file and bin-annotations are summarised in `vambbins_aggregated_annotation.txt`. 
 ```bash
-python mag_annotation/scripts/run_RF.py contigs.fna.gz vamb/clusters.tsv annotations resultdir
+run_RF.py contigs.fna.gz vamb/clusters.tsv annotations resultdir
 
 ls resultsidr
 resultdir/vambbins_aggregated_annotation.txt

phamb/__init__.py

@@ -0,0 +1,8 @@
+
+
+"""Phamb - Phages from metagenomic binning
+Documentation: https://github.com/RasmussenLab/phamb
+"""
+
+__licence__ = 'MIT'
+__version__ = (1, 0, 1)

…mag_annotation/dbs/RF_model.python39.sav phamb/dbs/RF_model.python39.savworkflows/mag_annotation/dbs/RF_model.python39.sav renamed to phamb/dbs/RF_model.python39.sav workflows/mag_annotation/dbs/RF_model.python39.sav renamed to phamb/dbs/RF_model.python39.sav

@@ -1,11 +1,14 @@
 #!/usr/bin/python
 import sys
 import argparse
-import vambtools as _vambtools
-import run_RF_modules
+#import vambtools as _vambtools
+from phamb import vambtools as _vambtools
+from phamb import run_RF_modules
+#import run_RF_modules
 import collections as _collections
 import os
 import numpy as _np
+from pathlib import Path
 
 
 parser = argparse.ArgumentParser(
@@ -185,7 +188,8 @@ def _run_RF_model(cls,RF_model,genome_order, sparse_df):
 
         print('Loading Model and annotation table')
         trained_model = joblib.load(RF_model)
-
+        trained_model.n_estimators = 300
+        trained_model.max_features = 'sqrt'
         predicted_genome_labels = trained_model.predict(sparse_df)
         prediction_probabilities = trained_model.predict_proba(sparse_df)
         predicted_genome_labels = [label.lower() for label in list(predicted_genome_labels) ] 
@@ -226,7 +230,7 @@ def _run_RF_model(cls,RF_model,genome_order, sparse_df):
 
     viral_annotation = run_RF_modules.Viral_annotation(annotation_files=viral_annotation_files,genomes=reference)
 
-    rf_model_file = 'mag_annotation/dbs/RF_model.python39.sav'
+    rf_model_file = Path(__file__).parent / "dbs/RF_model.python39.sav"
     RF_results = RF_model(rf_model_file,  genomes = viral_annotation.genomes)
 
     bins = {binname:clusters[binname] for binname in RF_results.RF_non_bacteria}

workflows/mag_annotation/dbs/RF_model.sav phamb/dbs/RF_model.savworkflows/mag_annotation/dbs/RF_model.sav renamed to phamb/dbs/RF_model.sav

@@ -1,6 +1,8 @@
 #!/bin/python
 '''Helper modules'''
-import vambtools as _vambtools
+#import vambtools as _vambtools
+from phamb import vambtools as _vambtools
+
 import collections as _collections
 import os
 import numpy as _np

…rkflows/mag_annotation/scripts/run_RF.py phamb/run_RF.pyworkflows/mag_annotation/scripts/run_RF.py renamed to phamb/run_RF.py workflows/mag_annotation/scripts/run_RF.py renamed to phamb/run_RF.py

@@ -0,0 +1,25 @@
+import sys
+from setuptools import setup, find_packages
+from setuptools import Extension
+import os
+
+SETUP_METADATA = \
+               {
+    "name": "phamb",
+    "description": "Phages from metagenomic binning",
+    "url": "https://github.com/RasmussenLab/phamb",
+    "version": "1.0.1",
+    "license": "MIT",
+    "packages": ['phamb'],
+    "package_data": {'phamb': ['dbs/RF_model.python39.sav']},
+    "python_requires": ">=3.9",
+    "install_requires": ["scikit-learn==1.0.2"],
+    "classifiers":[
+        "Programming Language :: Python :: 3",
+        "License :: OSI Approved :: MIT License",
+        "Operating System :: OS Independent",
+    ],
+    "scripts":['phamb/run_RF.py','phamb/split_contigs.py','phamb/vambtools.py','phamb/run_RF_modules.py']
+    }
+
+setup(**SETUP_METADATA)