kevinid / molecule_generator Goto Github PK

Code for "Multi-Objective De Novo Drug Design with Conditional Graph Generative Model" (https://arxiv.org/abs/1801.07299)

License: Apache License 2.0

Python 5.76% Jupyter Notebook 94.12% Dockerfile 0.12% Shell 0.01%

molecule_generator's Introduction

Conditional Molecule Generator

This repository contains the source code and data sets for the graph based molecule generator discussed in the article "Multi-Objective De Novo Drug Design with Conditional Graph Generative Model" (https://arxiv.org/abs/1801.07299).

Briefly speaking, we used conditional graph convolution to structure the generative model. The properties of output molecules can then be controlled using the conditional code.

Requirements

This repo is built using Python 2.7, and utilizes the following packages:

MXNet == 1.3.1
RDKit == 2018.03.3
Numpy
Scikit-learn (for the predictive model)

To ease the installation process, please use the dockerfile environment defined in the Dockerfile.

Quick start

Project structure

train.py: main training script.
mx_mg: package for the molecule generative model:
- data: packages for data processing workflows:
  - conditionals.py: callables used to generate the conditional codes for molecules
  - data_struct.py: defines atom types and bond types
  - dataloaders.py , datasets.py and samplers.py: data loading logics
  - utils.py: utility functions
- models: library for graph generative models
  - modules.py: define modules (or blocks) such as graph convolution
  - networks.py: define networks (MolMP, MolRNN and CMolRNN)
  - functions.py: autograd.Function objects and operations
- builders.py: utilities for building molecules using generative models
rdkit_contrib: functions used to calculate QED and SAscore (for older version of rdkit)
example.ipynb: tutorial

Usage

To train the model, first unpackdatasets.tar.gz (download here) to the current directory, and call:

./train.py {molmp|molrnn|scaffold|prop|kinase} path/to/output

Where {molmp|molrnn|scaffold|prop|kinase} are model types, and path/to/output is the directory where you want to save the model's checkpoint file and log files. The following call:

./train.py {molmp|molrnn|scaffold|prop|kinase} -h

gives help for each model type.

For any questions | problems | criticisms | ...

Please contact me. Email: [email protected] or [email protected]

molecule_generator's People

Contributors

Stargazers

Watchers

Forkers

tonydeep chennnnnyize-zz jin70908469 wllllg aspirincode abhik1368 cheminfopku rkakamilan amoliu minghao2016 champagnelee thegodone gkxiao xxffliu csc19960608 zcrwind leelasdsi amazingang jdhorwood deep-scaffold bbyun28 lxlsu akulashray yulanl22 yotta-private swyang-lab mshik cuidachao huzhyn sailfish009 psyche11 wangdi2014 ifyoungnet huangliang0828 rnaimehaom c10h m-hakmi gaoshan2006 asyaorlova gerard-ompad

molecule_generator's Issues

Missing model.json and traner.json file for tf_mg branch

Unable to find model.json and trainer.json files for tensorflow mg branch.

How do I run the tensorflow version of the code?

I am trying to work my way through the tf branch of the code. How do I run the branch using tensorflow?

There are some RDkit errors when I run the mxnet version code with pretrained models

When I run the example codes as follows

# loading models
mol_mp = builders.Vanilla_Builder('ckpt/molmp_0/', gpu_id=0)
res = mol_mp.sample(100)

I got the error info as follows:

[09:24:17] Can't kekulize mol.  Unkekulized atoms: 1 2 3 5 7 8 9 11 12 13 17

RDKit ERROR: [09:24:17] Can't kekulize mol.  Unkekulized atoms: 1 2 3 5 7 8 9 11 12 13 17
RDKit ERROR: 
[09:24:17] Can't kekulize mol.  Unkekulized atoms: 3 5 7

RDKit ERROR: [09:24:17] Can't kekulize mol.  Unkekulized atoms: 3 5 7
RDKit ERROR: 
[09:24:22] Can't kekulize mol.  Unkekulized atoms: 3 4 5 6 8 9 10 11 12 13 27

RDKit ERROR: [09:24:22] Can't kekulize mol.  Unkekulized atoms: 3 4 5 6 8 9 10 11 12 13 27
RDKit ERROR: 
[09:24:22] Explicit valence for atom # 38 N, 4, is greater than permitted
RDKit ERROR: [09:24:22] Explicit valence for atom # 38 N, 4, is greater than permitted
98

Is this a normal phenomenon? Many thanks.

Where can I find your additional files which were mentioned in your paper?

Additional Files

Additional file 1 — Supplementary Text
Containing additional information about the model architecture and
implementation details of experiments.

Additional file 2 — Supplementary Figures
Containing supplementary figures.

Trying to run examples.ipynb on Colab - serializing/multiprocessing issues

Hi! Thanks for providing the source code. I've been trying to get the mx_net implementation to work on an M1 Macbook Pro on Google Colab, since there were quite some issues with the MXNet + Cuda integration locally. I've made the neccesary changes where needed, adapted the imports of the files, but when trying to run the training code, it gives the following error in Colab:

!python3 train.py scaffold /content/molecule_generator

Traceback (most recent call last):
File "/content/molecule_generator/train.py", line 535, in
_engine(**params)
File "/content/molecule_generator/train.py", line 146, in _engine
inputs = [next(it_train) for _ in range(len(gpu_ids))]
File "/content/molecule_generator/train.py", line 146, in
inputs = [next(it_train) for _ in range(len(gpu_ids))]
File "/usr/local/lib/python3.10/dist-packages/mxnet/gluon/data/dataloader.py", line 689, in iter
for item in t:
File "/usr/local/lib/python3.10/dist-packages/mxnet/gluon/data/dataloader.py", line 484, in next
batch = pickle.loads(ret.get(self._timeout))
File "/usr/lib/python3.10/multiprocessing/pool.py", line 774, in get
raise self._value
File "/usr/lib/python3.10/multiprocessing/pool.py", line 540, in _handle_tasks
put(task)
File "/usr/lib/python3.10/multiprocessing/connection.py", line 206, in send
self._send_bytes(_ForkingPickler.dumps(obj))
File "/usr/lib/python3.10/multiprocessing/reduction.py", line 51, in dumps
cls(buf, protocol).dump(obj)
File "/usr/lib/python3.10/multiprocessing/pool.py", line 643, in reduce
raise NotImplementedError(
NotImplementedError: pool objects cannot be passed between processes or pickled

I was wondering if anyone else had a similar error, and whether someone manages to fix this one.

Is [#6,#8,#16][#6](=O)O[#6] an alert？

This smart is in the structural alert list. This means that all esters are unwanted structures. Is that so, or the '#6' should be removed from the first square bracket. At this time, it means that compounds like carbonates are containning the alert structures. Thanks!

how to get Bemis-Murcko scaffolds and ring assemblies

how can i get Bemis-Murcko scaffolds and ring assemblies in my own dataset？

When I clone your tf branch, I still got the same package as mxnet version

I only found tf_mg, but your example file still uses mxnet

running errors

I run exmples.ipynb, but received error message like this
No module named 'utils'

I found file utils.py at /data/ folder. Can you tell me why?
I am using python 3.5

Thanks

I can not find the file"datasets/atom_types.txt"

Hi, I can not find the file"datasets/atom_types.txt", would you mind uploading this file to your program?

Error while running the Python3 version of this code

I converted the entire code into Python-3.7 using 2to3 package provided by Python. While running the converted code, I got following error:
AttributeError: Can't pickle local object '_engine.<locals>.<lambda>'

Complete error is as follows:

$ ./train.py molmp logs
Traceback (most recent call last):
  File "./train.py", line 551, in <module>
    _engine(**params)
  File "./train.py", line 133, in _engine
    inputs = [next(it_train) for _ in range(len(gpu_ids))]
  File "./train.py", line 133, in <listcomp>
    inputs = [next(it_train) for _ in range(len(gpu_ids))]
  File "/home/iit/anaconda3/envs/kevinid_experiments/lib/python3.7/site-packages/mxnet/gluon/data/dataloader.py", line 505, in __next__
    batch = pickle.loads(ret.get(self._timeout))
  File "/home/iit/anaconda3/envs/kevinid_experiments/lib/python3.7/multiprocessing/pool.py", line 657, in get
    raise self._value
  File "/home/iit/anaconda3/envs/kevinid_experiments/lib/python3.7/multiprocessing/pool.py", line 431, in _handle_tasks
    put(task)
  File "/home/iit/anaconda3/envs/kevinid_experiments/lib/python3.7/multiprocessing/connection.py", line 206, in send
    self._send_bytes(_ForkingPickler.dumps(obj))
  File "/home/iit/anaconda3/envs/kevinid_experiments/lib/python3.7/multiprocessing/reduction.py", line 51, in dumps
    cls(buf, protocol).dump(obj)
AttributeError: Can't pickle local object '_engine.<locals>.<lambda>'

Can someone please help me resolve this issue?