Merge remote-tracking branch 'remotes/origin/master' into MPNN

sudo: required

dist: trusty

install:

- if [[ "$TRAVIS_PYTHON_VERSION" == "2.7" ]]; then wget https://repo.continuum.io/archive/Anaconda2-4.3.0-Linux-x86_64.sh

  -O anaconda.sh; else wget https://repo.continuum.io/archive/Anaconda3-4.3.0-Linux-x86_64.sh

  -O anaconda.sh; fi

- wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O anaconda.sh;

- export python_version=$TRAVIS_PYTHON_VERSION

- bash anaconda.sh -b -p $HOME/anaconda

- export PATH="$HOME/anaconda/bin:$PATH"

- conda config --add channels http://conda.binstar.org/omnia

- bash scripts/install_deepchem_conda.sh deepchem

- source activate deepchem

- pip install yapf==0.16.2

- pip install yapf==0.17.0

- pip install coveralls

- python setup.py install

script:

DeepChem uses [yapf](https://github.com/google/yapf) to autoformat code.

``` bash

pip install yapf==0.16.2

pip install yapf==0.17.0

cd <git_root>

yapf -i <python_files changed>

```

DeepChem is supported by a number of corporate partners who use DeepChem to solve interesting problems.

### Schrödinger

[![Schödinger](https://github.com/deepchem/deepchem/raw/master/docs/_static/schrodinger_logo.png)](https://www.schrodinger.com/)

[![Schödinger](https://github.com/deepchem/deepchem/blob/master/docs/source/_static/schrodinger_logo.png)](https://www.schrodinger.com/)

> DeepChem has transformed how we think about building QSAR and QSPR models when very large data sets are available; and we are actively using DeepChem to investigate how to best combine the power of deep learning with next generation physics-based scoring methods.

### DeepCrystal

<img src="https://raw.githubusercontent.com/deepchem/deepchem/master/docs/_static/deep_crystal_logo.png" alt="DeepCrystal Logo" height=150px/>

<img src="https://github.com/deepchem/deepchem/blob/master/docs/source/_static/deep_crystal_logo.png" alt="DeepCrystal Logo" height=150px/>

> DeepCrystal was an early adopter of DeepChem, which we now rely on to abstract away some of the hardest pieces of deep learning in drug discovery. By open sourcing these efficient implementations of chemically / biologically aware deep-learning systems, DeepChem puts the latest research into the hands of the scientists that need it, materially pushing forward the field of in-silico drug discovery in the process.

## Version

1.1.0

1.2.0

__license__ = "MIT"

import sys

from deepchem.utils.dependencies import mdtraj as md

import unittest

import os

import numpy as np

    box2 = ((1, 3), (1, 3), (1, 3))

    mapping = {box1: [1, 2, 3, 4], box2: [1, 2, 3, 4, 5]}

    boxes = [box1, box2]

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(mapping,

                                                                     boxes)

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(

        mapping, boxes)

    print("merged_boxes")

    print(merged_boxes)

    assert len(merged_boxes) == 1

    box2 = ((1, 2), (1, 2), (1, 2))

    mapping = {box1: [1, 2, 3, 4, 5, 6], box2: [1, 2, 3, 4]}

    boxes = [box1, box2]

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(mapping,

                                                                     boxes)

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(

        mapping, boxes)

    print("merged_boxes")

    print(merged_boxes)

    assert len(merged_boxes) == 1

        box2: [1, 2, 3, 4],

        box3: [1, 2, 3, 4, 5]

    }

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(mapping,

                                                                     boxes)

    merged_boxes, _ = dc.dock.binding_pocket.merge_overlapping_boxes(

        mapping, boxes)

    print("merged_boxes")

    print(merged_boxes)

    assert len(merged_boxes) == 1

    protein_file = os.path.join(current_dir, "1jld_protein.pdb")

    ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")

    import mdtraj as md

    protein = md.load(protein_file)

    finder = dc.dock.ConvexHullPocketFinder()

    all_pockets = finder.find_all_pockets(protein_file)

    pockets, pocket_atoms_map, pocket_coords = finder.find_pockets(protein_file,

                                                                   ligand_file)

    pockets, pocket_atoms_map, pocket_coords = finder.find_pockets(

        protein_file, ligand_file)

    # Test that every atom in pocket maps exists

    n_protein_atoms = protein.xyz.shape[1]

    print("protein.xyz.shape")

    protein_file = os.path.join(current_dir, "1jld_protein.pdb")

    ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")

    import mdtraj as md

    protein = md.load(protein_file)

    finder = dc.dock.RFConvexHullPocketFinder()

    pockets, pocket_atoms_map, pocket_coords = finder.find_pockets(protein_file,

                                                                   ligand_file)

    pockets, pocket_atoms_map, pocket_coords = finder.find_pockets(

        protein_file, ligand_file)

    # Test that every atom in pocket maps exists

    n_protein_atoms = protein.xyz.shape[1]

    print("protein.xyz.shape")

__license__ = "LGPL v2.1+"

import numpy as np

from deepchem.utils.dependencies import mdtraj

from deepchem.feat import Featurizer

from deepchem.feat import ComplexFeaturizer

from deepchem.utils import rdkit_util, pad_array

                          periodic_box_size):

  """Computes a neighbor list from atom coordinates."""

  N = coords.shape[0]

  import mdtraj

  traj = mdtraj.Trajectory(coords.reshape((1, N, 3)), None)

  box_size = None

  if periodic_box_size is not None:

      dist = np.linalg.norm(delta, axis=1)

      sorted_neighbors = list(zip(dist, neighbors[i]))

      sorted_neighbors.sort()

      neighbor_list[

          i] = [sorted_neighbors[j][1] for j in range(max_num_neighbors)]

      neighbor_list[i] = [

          sorted_neighbors[j][1] for j in range(max_num_neighbors)

      ]

    else:

      neighbor_list[i] = list(neighbors[i])

  return neighbor_list