Merge pull request #32 from LokiLuciferase/feature/documentation

Installation and Usage documentation

README.rst

@@ -3,7 +3,7 @@ phenotrex
 =========
 
 
-.. image:: https://img.shields.io/pypi/v/phenotrex.svg
+.. image:: https://badge.fury.io/py/phenotrex.svg
    :target: https://pypi.python.org/pypi/phenotrex
    :alt: PyPI
 
@@ -30,5 +30,38 @@ phenotrex
 
 End-to-end Microbial Phenotypic Trait Prediction.
 
-* Supported platforms: Linux, MacOS, Windows
-* Free software: GPLv3 license
+Installation
+------------
+
+.. code-block::
+
+    $ pip install phenotrex[fasta]
+
+Usage
+-----
+
+Phenotrex is a component of the `PhenDB`_ web server, which performs phenotypic trait prediction on
+user-uploaded metagenomic bins. To try out phenotrex with PhenDB's pre-trained and curated set of
+trait models, genomes may thus simply be `submitted to PhenDB`_.
+
+Basic Usage
+~~~~~~~~~~~
+To use a trained phenotrex model ``MY_TRAIT.pkl`` for prediction of a phenotypic trait with a
+given genome ``genome.fna``:
+
+.. code-block::
+
+    $ phenotrex predict --classifier MY_TRAIT.pkl genome.fna > predictions.tsv
+
+
+This yields a tabular file containing a prediction regarding the presence of the trait (YES or NO),
+as well as a confidence value the model ascribes to this prediction, ranging from 0.5 to 1.0.
+
+Advanced Usage
+~~~~~~~~~~~~~~
+For training, evaluation and explanation of phenotrex models on user data, please refer to the
+full usage tutorial `here`_.
+
+.. _PhenDB: https://www.phendb.org/
+.. _submitted to PhenDB: https://phen.csb.univie.ac.at/phendb/
+.. _here: https://phenotrex.readthedocs.io/en/latest/usage.html

docs/installation.rst

@@ -8,17 +8,28 @@ Installation
 Stable release
 --------------
 
-To install phenotrex, run this command in your terminal:
+For a full installation of phenotrex, run this command in your terminal:
 
 .. code-block:: console
 
-    $ pip install phenotrex
+    $ pip install phenotrex[fasta]
+
+Note that this command installs large dependencies (`pytorch`_, `deepnog`_) required for
+transforming FASTA files into phenotrex input features at runtime.
+
+If this capability is not needed (for example because feature files
+have been pre-created),
+installation size can be significantly reduced by running instead:
 
-This is the preferred method to install phenotrex, as it will always install the most recent stable release.
+.. code-block:: console
+
+    $ pip install phenotrex
 
 If you don't have `pip`_ installed, this `Python installation guide`_ can guide
 you through the process.
 
+.. _pytorch: https://pytorch.org/
+.. _deepnog: https://github.com/univieCUBE/deepnog
 .. _pip: https://pip.pypa.io
 .. _Python installation guide: http://docs.python-guide.org/en/latest/starting/installation/
 
@@ -38,13 +49,13 @@ Or download the `tarball`_:
 
 .. code-block:: console
 
-    $ curl  -OL https://github.com/univieCUBE/phenotrex/tarball/master
+    $ curl -OL https://github.com/univieCUBE/phenotrex/tarball/master
 
 Once you have a copy of the source, you can install it with:
 
 .. code-block:: console
 
-    $ make install
+    $ make full-install
 
 
 .. _Github repo: https://github.com/univieCUBE/phenotrex