Develop (#19)

* Add github action to publish package to PyPi

* Feature GitHub action conda build (#1)

* Add conda github action that builds the package with conda. It is triggered on pull request.

* Add pyro-ppl and pip as conda dependency. Trigger setup automatically via conda yaml. Split steps in conda build github action.

* Less strict in the setup.py dependencies for torch. Small change in the installation instructions in the documentation.

* Feature - Add basic unit tests to cover activation_functions.py (#3)

* adding first unit tests

* Format conda-build YAML file

* Update action name

* Fix path to __version__

* Test commit - Trigger workflow

* Revert dummy change

Co-authored-by: Mathias Rechenmann <[email protected]>

* Feature clean helper functions (#4)

* delete _helper_functions.py

* delete _hyperparams_optimization.py

* delete plot_every_epoch function

* delete import _helper_functions

* delete gpyopt dependency

* edit only README

* Add pylint github action

* Remove pypi github action (#9)

* support CPU, modify README (#10)

* support CPU, modify README

* shorten the import, add model to __init__

* More lenient scar installation specifications (#8)

* Split scAR.yml to scAR-gpu.yaml and scAR-cpu.yaml.
* More lenient installation specifications.
* Bump version to 0.2.0

* fix a typo, reorganise __init__.py

* fix bugs

* add synthetic data for integration test (#13)

* add synthetic data for integration test

* change paths

* change paths

* Remove torchaudio. Bump version to 0.2.2 (#14)

* Remove torchaudio
* Bump version to 0.2.2

* Update readme (#16)

* update README

* Black github action (#17)

Addition of black github action that runs on every push and every pull request. It shows in the stdout all the changes that need to be made (--diff), but returns exit code 0, even if errors are observed.

* Addition of integration test (#18)

* Add integration test as unit test

* update version

Co-authored-by: Gypas, Foivos <[email protected]>
Co-authored-by: Foivos Gypas <[email protected]>
Co-authored-by: Mathias Rechenmann <[email protected]>

.github/workflows/black.yaml

@@ -0,0 +1,14 @@
+name: Black lint
+
+on: [push, pull_request]
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - uses: psf/black@stable
+        with:
+          options: "--diff --verbose --line-length 127"
+          src: "./scAR"
+          version: "22.3.0"

README.md

@@ -1,10 +1,8 @@
 # scAR  
 
-[![scAR](https://img.shields.io/badge/scAR-005AF0?style=for-the-badge&logo=dependabot&logoColor=white.svg)](https://github.com/Novartis/scAR)
-![single-cell omics](https://img.shields.io/badge/single_cell_omics-005AF0?style=for-the-badge.svg)
-![machine learning](https://img.shields.io/badge/machine_learning-005AF0?style=for-the-badge.svg)
-![variational autoencoders](https://img.shields.io/badge/variational_autoencoders-005AF0?style=for-the-badge.svg)
-![denoising](https://img.shields.io/badge/denoising-005AF0?style=for-the-badge.svg)
+[![scAR](https://anaconda.org/bioconda/scar/badges/version.svg)](https://anaconda.org/bioconda/scar)
+[![Stars](https://img.shields.io/github/stars/Novartis/scar?logo=GitHub&color=red)](https://github.com/Novartis/scAR)
+[![Downloads](https://anaconda.org/bioconda/scar/badges/downloads.svg)](https://anaconda.org/bioconda/scar/files)
 
 **scAR** (single cell Ambient Remover) is a package for denoising multiple single cell omics data. It can be used for multiple tasks, such as, **sgRNA assignment** for scCRISPRseq, **identity barcode assignment** for cell indexing, **protein denoising** for CITE-seq, **mRNA denoising** for scRNAseq, and etc... It is built using probabilistic deep learning, illustrated as follows:
 
@@ -22,7 +20,25 @@
 
 ## Installation
 
-Clone this repository,
+#### Conda
+1, Install [conda](https://docs.conda.io/projects/conda/en/latest/user-guide/install/index.html)  
+2, Create conda environment
+```sh
+$ conda create -n scAR_env
+```
+
+3, Activate conda environment
+```sh
+$ conda activate scAR_env
+```  
+
+4, Install scar
+```sh
+$ conda install -c bioconda scar
+```
+
+#### Git+pip
+Alternatively, clone this repository,
 
 ```sh
 $ git clone https://github.com/Novartis/scAR.git
@@ -61,7 +77,7 @@ There are two ways to run scAR.
 1) Use scAR API if you are Python users
 
 ```sh
->>> from scAR import model
+>>> from scar import model
 >>> scarObj = model(adata.X.to_df(), empty_profile)
 >>> scarObj.train()
 >>> scarObj.inference()

scAR/main/__init__.py

@@ -1,2 +1,3 @@
 # -*- coding: utf-8 -*-
-__version__ = '0.2.2'
+
+__version__ = '0.2.3'

scAR/test/test_scar.py

@@ -0,0 +1,23 @@
+import pandas as pd
+from scAR import model
+import unittest
+
+class ScarIntegration(unittest.TestCase):
+    """
+    Test activation_functions.py functions.
+    """    
+
+    def test_scar(self):
+        raw_count = pd.read_pickle("scAR/test/raw_counts.pickle")
+        empty_profile=pd.read_pickle("scAR/test/ambient_profile.pickle")
+        expected_output=pd.read_pickle("scAR/test/output_assignment.pickle")
+
+        scarObj = model(
+            raw_count=raw_count.values, empty_profile=empty_profile, scRNAseq_tech="CROPseq"
+        )
+
+        scarObj.train(epochs=40, batch_size=64)
+
+        scarObj.inference()
+
+        self.assertTrue(scarObj.feature_assignment.equals(expected_output))