Merge branch 'master' of github.com:gnina/gnina

README.md

@@ -8,18 +8,30 @@ Help
 
 Citation
 ========
-If you find gnina useful, please cite our paper (hopefully more to come):  
-**Protein–Ligand Scoring with Convolutional Neural Networks**  
+If you find gnina useful, please cite our paper(s):  
+
+**Protein–Ligand Scoring with Convolutional Neural Networks**  (Primary citation)
 M Ragoza, J Hochuli, E Idrobo, J Sunseri, DR Koes. *J. Chem. Inf. Model*, 2017  
-[link](http://pubs.acs.org/doi/full/10.1021/acs.jcim.6b00740) [arXiv version](https://arxiv.org/abs/1612.02751)
+[link](http://pubs.acs.org/doi/full/10.1021/acs.jcim.6b00740) [arXiv](https://arxiv.org/abs/1612.02751)  
+
+**Ligand pose optimization with atomic grid-based convolutional neural networks**
+M Ragoza, L Turner, DR Koes. *Machine Learning for Molecules and Materials NIPS 2017 Workshop*, 2017
+[arXiv](https://arxiv.org/abs/1710.07400)  
 
+**Visualizing convolutional neural network protein-ligand scoring**
+J Hochuli, A Helbling, T Skaist, M Ragoza, DR Koes.  *Journal of Molecular Graphics and Modelling*, 2018
+[link](https://www.sciencedirect.com/science/article/pii/S1093326318301670) [arXiv](https://arxiv.org/abs/1803.02398)
+
+**Convolutional neural network scoring and minimization in the D3R 2017 community challenge**
+J Sunseri, JE King, PG Francoeur, DR Koes.  *Journal of computer-aided molecular design*, 2018
+[link](https://link.springer.com/article/10.1007/s10822-018-0133-y) [PubMed](https://www.ncbi.nlm.nih.gov/pubmed/29992528)
 
 Installation
 ============
 
 ### Ubuntu 16.04
 ```
-apt-get install build-essential git wget libopenbabel-dev libboost-all-dev libeigen3-dev libgoogle-glog-dev libprotobuf-dev protobuf-compiler libhdf5-serial-dev libatlas-base-dev python-dev cmake librdkit-dev python-numpy
+apt-get install build-essential git wget libopenbabel-dev libboost-all-dev libeigen3-dev libgoogle-glog-dev libprotobuf-dev protobuf-compiler libhdf5-serial-dev libatlas-base-dev python-dev cmake librdkit-dev python-numpy python-pip
 ```
 
 [Follow NVIDIA's instructions](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/#axzz4TWipdwX1) to install the latest version of CUDA.  *Note* we are in the process of transitioning to CUDA 9.1.
@@ -36,6 +48,8 @@ make install
 ```
 # 
 
+Note you the scripts provided in `gnina/scripts` have additional python dependencies that must be installed. 
+
 ### CentOS 7
 
 The program will not build in a computer with a gpu with computer capability < 3.5 unless
@@ -190,7 +204,8 @@ The provided models are templated with `TRAINFILE` and `TESTFILE` arguments, whi
 actual files.  The `train.py` script can be called with a model and a prefix for testing and training files:
 
 ```
-train.py -m models/refmodel3/refmodel3.model -p models/data/csar/all
+cd models/refmodel3
+train.py -m refmodel3.model -p ../data/csar/all -d ../data/csar
 ```
 
 This will perform cross-validation using the `alltrain[0-2].types` and `alltest[0-2].types` files.