Fix install

README.md

@@ -13,17 +13,48 @@ SCI is a program to identify sub-compartments from HiC data. SCI utilizes graph
 * python 2.7
 
 **Python Libraries**
-* [scikit-learn] >=0.19.0 
+* [scikit-learn] >=0.19.0, <= 0.20
 * [Numpy] >= 1.15
 * [tqdm]>=4.24
 
+`Scikit-learn 0.20 was the last version to support Python 2.7 and Python 3.4. Scikit-learn now requires Python 3.5 or newer.`  
+
 **C++ libraries**
 * [GSL]
 
+### using Conda  
+
+use conda-forge channel  
+
+```sh
+conda config --add channels conda-forge
+
+# 
+conda create -n sci_env python=2.7
+
+# once created
+conda activate sci_env
+
+# install own compilers
+conda install compilers
+
+# install GSL
+conda install gsl
+
+# save your env
+mkdir -p envs
+conda env export >envs/sci_env.yml
+
+```
+
 ## Installation 
 
 ```sh
-$ python setup.py
+$ python setup.py install
+
+# clean
+$ python setup.py clean
+
 ```
 ## Input format
 
@@ -76,6 +107,9 @@ SCI output sub-compartments annotation into BED format with the following fields
 
 
 ## Test run
+
+`IS MISSING`
+
 To preform test run for SCI please follow the following steps:
 1. Go the Input_sample directory
 ```sh
@@ -99,4 +133,4 @@ $ python sci.py -n test -f Input_sample/SCI_input.txt -r 100000 -g chromosome_si
 [scikit-learn]: http://scikit-learn.org/stable/
 [Numpy]: http://www.numpy.org/
 [tqdm]: https://pypi.org/project/tqdm/
-[GSL]:  http://www.gnu.org/software/gsl/
+[GSL]:  http://www.gnu.org/software/gsl/

requirements.txt

@@ -0,0 +1,8 @@
+# Scikit-learn 0.20 was the last version to support Python 2.7 and Python 3.4. Scikit-learn now requires Python 3.5 or newer.
+#scikit-learn >= 0.18
+scikit-learn > 0.18, < 0.21
+# last numpy release tosupport Python 2.7: see https://docs.scipy.org/doc/numpy/release.html#numpy-1-16-0-release-notes
+Numpy <=1.16.0
+# The last SciPy version to do so is SciPy 1.2.x: https://www.scipy.org/scipylib/faq.html#do-numpy-and-scipytill-support-python-2-7
+scipy <=1.2
+tqdm

setup.py

@@ -8,16 +8,26 @@
 VERSION = '0.1.0'
 
 
+CONDA_PATH =  os.getenv("CONDA_PREFIX")
+INCPATH = os.path.join(CONDA_PATH,"include")
+LIBPATH=os.path.join(CONDA_PATH,"lib")
+
+
+LIBS="-lgsl -lgslcblas -lpthread -L{}".format(LIBPATH)
+
 def compile_line():
-    compile_command = ("g++ LINE/line.cpp -o LINE/line"
-                       "-lgsl -lgslcblas -lpthread")
+    compile_command = ("g++ LINE/line.cpp -I{}  -o LINE/line {}".format(INCPATH,LIBS))
     return_code = call(compile_command.split())
     if return_code != 0:
         sys.exit(("LINE compilation has failed."
                   "Please make sure to install GSL library"))
 
 compile_line()
 
+## Important notes to requirements:
+# 1. Last numpy release tosupport Python 2.7: see https://docs.scipy.org/doc/numpy/release.html#numpy-1-16-0-release-notes
+# 2. The last SciPy version to do so is SciPy 1.2.x: https://www.scipy.org/scipylib/faq.html#do-numpy-and-scipytill-support-python-2-7 
+# 3. Scikit-learn 0.20 was the last version to support Python 2.7 and Python 3.4. Scikit-learn now requires Python 3.5 or newer.
 setup(
 
     name="sci",
@@ -26,8 +36,9 @@ def compile_line():
     author_email='[email protected]',
     packages=find_packages(),
     install_requires=[
-        'numpy',
-        'scikit-learn',
+        'numpy<=1.16.0',
+        'scipy<=1.2',
+        'scikit-learn>=0.19,<=0.20',
         'tqdm'
     ],