DeepChem aims to provide a high quality open-source toolchain that democratizes the use of deep-learning in drug discovery, materials science, and quantum chemistry. DeepChem is a package developed by the Pande group at Stanford and originally created by Bharath Ramsundar.

• Requirements
• Installation from Source
• FAQ
• Getting Started
  • Input Formats
  • Data Featurization
  • Performances
• Contributing to DeepChem
  • Code Style Guidelines
  • Documentation Style Guidelines
  • Gitter
• DeepChem Publications
• Examples
• About Us

• openbabel
• pandas
• rdkit
• boost
• joblib
• sklearn
• numpy
• six
• mdtraj
• tensorflow

Installation from source is the only currently supported format. deepchem currently supports both Python 2.7 and Python 3.5, but is not supported on any OS'es except 64 bit linux. Please make sure you follow the directions below precisely. While you may already have system versions of some of these packages, there is no guarantee that deepchem will work with alternate versions than those specified below.

1. Download the 64-bit Python 2.7 or Python 3.5 versions of Anaconda for linux here.

   Follow the installation instructions

2. openbabel

   conda install -c omnia openbabel=2.4.0

3. rdkit

   conda install -c rdkit rdkit

4. joblib

   conda install joblib 

5. six

   pip install six

6. mdtraj

   conda install -c omnia mdtraj

7. tensorflow: Installing tensorflow on older versions of Linux (which have glibc < 2.17) can be very challenging. For these older Linux versions, contact your local sysadmin to work out a custom installation. If your version of Linux is recent, then the following command will work:

   pip install tensorflow-gpu==0.12.1
   

8. deepchem: Clone the deepchem github repo:

   git clone https://github.com/deepchem/deepchem.git

   cd into the deepchem directory and execute

   python setup.py install

9. To run test suite, install nosetests:

   pip install nose

   Make sure that the correct version of nosetests is active by running

   which nosetests 

   You might need to uninstall a system install of nosetests if there is a conflict.

10. If installation has been successful, all tests in test suite should pass:

    nosetests -v deepchem --nologcapture 

    Note that the full test-suite uses up a fair amount of memory. Try running tests for one submodule at a time if memory proves an issue.

Alternatively, you can install deepchem in a new conda environment using the conda commands in scripts/install_deepchem_conda.sh

bash scripts/install_deepchem_conda.sh deepchem
pip install tensorflow-gpu==0.12.1                      # If you want GPU support
git clone https://github.com/deepchem/deepchem.git      # Clone deepchem source code from GitHub
cd deepchem
python setup.py install                                 # Manual install
nosetests -v deepchem --nologcapture                    # Run tests

This creates a new conda environment deepchem and installs in it the dependencies that are needed. To access it, use the source activate deepchem command. Check this link for more information about the benefits and usage of conda environments. Warning: Segmentation faults can still happen via this installation procedure.

1. Question: I'm seeing some failures in my test suite having to do with MKL Intel MKL FATAL ERROR: Cannot load libmkl_avx.so or libmkl_def.so.

   Answer: This is a general issue with the newest version of scikit-learn enabling MKL by default. This doesn't play well with many linux systems. See BVLC/caffe#3884 for discussions. The following seems to fix the issue

   conda install nomkl numpy scipy scikit-learn numexpr
   conda remove mkl mkl-service

2. Question: The test suite is core-dumping for me. What's up?

   [rbharath]$ nosetests -v deepchem --nologcapture
   Illegal instruction (core dumped)
   

   Answer: This is often due to openbabel issues on older linux systems. Open ipython and run the following

   In [1]: import openbabel as ob
   

   If you see a core-dump, then it's a sign there's an issue with your openbabel install. Try reinstalling openbabel from source for your machine.

The first step to getting started is looking at the examples in the examples/ directory. Try running some of these examples on your system and verify that the models train successfully. Afterwards, to apply deepchem to a new problem, try starting from one of the existing examples and modifying it step by step to work with your new use-case.

Accepted input formats for deepchem include csv, pkl.gz, and sdf files. For example, with a csv input, in order to build models, we expect the following columns to have entries for each row in the csv file.

1. A column containing SMILES strings [1].
2. A column containing an experimental measurement.
3. (Optional) A column containing a unique compound identifier.

Here's an example of a potential input file.

Here the "smiles" column contains the SMILES string, the "measured log solubility in mols per litre" contains the experimental measurement and "Compound ID" contains the unique compound identifier.

[2] Anderson, Eric, Gilman D. Veith, and David Weininger. "SMILES, a line notation and computerized interpreter for chemical structures." US Environmental Protection Agency, Environmental Research Laboratory, 1987.

Most machine learning algorithms require that input data form vectors. However, input data for drug-discovery datasets routinely come in the format of lists of molecules and associated experimental readouts. To transform lists of molecules into vectors, we need to subclasses of DeepChem loader class dc.data.DataLoader such as dc.data.CSVLoader or dc.data.SDFLoader. Users can subclass dc.data.DataLoader to load arbitrary file formats. All loaders must be passed a dc.feat.Featurizer object. DeepChem provides a number of different subclasses of dc.feat.Featurizer for convenience.

• Classification

Index splitting

Random splitting

Scaffold splitting

• Regression

• General features

Number of tasks and examples in the datasets

Time needed for benchmark test(~20h in total)

We actively encourage community contributions to DeepChem. The first place to start getting involved is by running our examples locally. Afterwards, we encourage contributors to give a shot to improving our documentation. While we take effort to provide good docs, there's plenty of room for improvement. All docs are hosted on Github, either in this README.md file, or in the docs/ directory.

Once you've got a sense of how the package works, we encourage the use of Github issues to discuss more complex changes, raise requests for new features or propose changes to the global architecture of DeepChem. Once consensus is reached on the issue, please submit a PR with proposed modifications. All contributed code to DeepChem will be reviewed by a member of the DeepChem team, so please make sure your code style and documentation style match our guidelines!

DeepChem uses yapf to autoformat code. We created a git pre-commit hook to make this process easier.

cp devtools/travis-ci/pre-commit .git/hooks
pip install yapf==0.16.0

DeepChem uses NumPy style documentation. Please follow these conventions when documenting code, since we use Sphinx+Napoleon to automatically generate docs on deepchem.io.

Join us on gitter at https://gitter.im/deepchem/Lobby. Probably the easiest place to ask simple questions or float requests for new features.

1. Computational Modeling of β-secretase 1 (BACE-1) Inhibitors using Ligand Based Approaches
2. Low Data Drug Discovery with One-shot Learning
3. MoleculeNet: A Benchmark for Molecular Machine Learning

DeepChem is a package by the Pande group at Stanford. DeepChem was originally created by Bharath Ramsundar, and has grown through the contributions of a number of undergraduate, graduate, and postdoctoral researchers working with the Pande lab.