Public Implementation of ChIRo from "Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations"

Related tags

Deep LearningChIRo
Overview

Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations

ScreenShot

This directory contains the model architectures and experimental setups used for ChIRo, SchNet, DimeNet++, and SphereNet on the four tasks considered in the preprint:

Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations

These four tasks are:

  1. Contrastive learning to cluster conformers of different stereoisomers in a learned latent space
  2. Classification of chiral centers as R/S
  3. Classification of the sign (+/-; l/d) of rotated circularly polarized light
  4. Ranking enantiomers by their docking scores in an enantiosensitive protein pocket.

The exact data splits used for tasks (1), (2), and (4) can be downloaded from:

https://figshare.com/s/e23be65a884ce7fc8543

See the appendix of "Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations" for details on how the datasets for task (3) were extracted and filtered from the commercial Reaxys database.


This directory is organized as follows:

  • Subdirectory model/ contains the implementation of ChIRo.

    • model/alpha_encoder.py contains the network architecture of ChIRo

    • model/embedding_functions.py contains the featurization of the input conformers (RDKit mol objects) for ChIRo.

    • model/datasets_samplers.py contains the Pytorch / Pytorch Geometric data samplers used for sampling conformers in each training batch.

    • model/train_functions.py and model/train_models.py contain supporting training/inference loops for each experiment with ChIRo.

    • model/optimization_functions.py contains the loss functions used in the experiments with ChIRo.

    • Subdirectory model/gnn_3D/ contains the implementations of SchNet, DimeNet++, and SphereNet used for each experiment.

      • model/gnn_3D/schnet.py contains the publicly available code for SchNet, with adaptations for readout.
      • model/gnn_3D/dimenet_pp.py contains the publicly available code for DimeNet++, with adaptations for readout.
      • model/gnn_3D/spherenet.py contains the publicly available code for SphereNet, with adaptations for readout.
      • model/gnn_3D/train_functions.py and model/gnn_3D/train_models.py contain the training/inference loops for each experiment with SchNet, DimeNet++, or SphereNet.
      • model/gnn_3D/optimization_functions.py contains the loss functions used in the experiments with SchNet, DimeNet++, or SphereNet.
  • Subdirectory params_files/ contains the hyperparameters used to define exact network initializations for ChIRo, SchNet, DimeNet++, and SphereNet for each experiment. The parameter .json files are specified with a random seed = 1, and the first fold of cross validation for the l/d classifcation task. For the experiments specified in the paper, we use random seeds = 1,2,3 when repeating experiments across three training/test trials.

  • Subdirectory training_scripts/ contains the python scripts to run each of the four experiments, for each of the four 3D models ChIRo, SchNet, DimeNet++, and SphereNet. Before running each experiment, move the corresponding training script to the parent directory.

  • Subdirectory hyperopt/ contains hyperparameter optimization scripts for ChIRo using Raytune.

  • Subdirectory experiment_analysis/ contains jupyter notebooks for analyzing results of each experiment.

  • Subdirectory paper_results/ contains the parameter files, model parameter dictionaries, and loss curves for each experiment reported in the paper.


To run each experiment, first create a conda environment with the following dependencies:

  • python = 3.8.6
  • pytorch = 1.7.0
  • torchaudio = 0.7.0
  • torchvision = 0.8.1
  • torch-geometric = 1.6.3
  • torch-cluster = 1.5.8
  • torch-scatter = 2.0.5
  • torch-sparce = 0.6.8
  • torch-spline-conv = 1.2.1
  • numpy = 1.19.2
  • pandas = 1.1.3
  • rdkit = 2020.09.4
  • scikit-learn = 0.23.2
  • matplotlib = 3.3.3
  • scipy = 1.5.2
  • sympy = 1.8
  • tqdm = 4.58.0

Then, download the datasets (with exact training/validation/test splits) from https://figshare.com/s/e23be65a884ce7fc8543 and place them in a new directory final_data_splits/

You may then run each experiment by calling:

python training_{experiment}_{model}.py params_files/params_{experiment}_{model}.json {path_to_results_directory}/

For instance, you can run the docking experiment for ChIRo with a random seed of 1 (editable in the params .json file) by calling:

python training_binary_ranking.py params_files/params_binary_ranking_ChIRo.json results_binary_ranking_ChIRo/

After training, this will create a results directory containing model checkpoints, best model parameter dictionaries, and results on the test set (if applicable).

Implementation of SE3-Transformers for Equivariant Self-Attention, in Pytorch.

SE3 Transformer - Pytorch Implementation of SE3-Transformers for Equivariant Self-Attention, in Pytorch. May be needed for replicating Alphafold2 resu

Phil Wang 207 Dec 23, 2022
Pytorch for Segmentation

Pytorch for Semantic Segmentation This repo has been deprecated currently and I will not maintain it. Meanwhile, I strongly recommend you can refer to

ycszen 411 Nov 22, 2022
Relative Positional Encoding for Transformers with Linear Complexity

Stochastic Positional Encoding (SPE) This is the source code repository for the ICML 2021 paper Relative Positional Encoding for Transformers with Lin

Antoine Liutkus 48 Nov 16, 2022
PyTorch Implementation of "Light Field Image Super-Resolution with Transformers"

LFT PyTorch implementation of "Light Field Image Super-Resolution with Transformers", arXiv 2021. [pdf]. Contributions: We make the first attempt to a

Squidward 62 Nov 28, 2022
This repo includes our code for evaluating and improving transferability in domain generalization (NeurIPS 2021)

Transferability for domain generalization This repo is for evaluating and improving transferability in domain generalization (NeurIPS 2021), based on

gordon 9 Nov 29, 2022
Official PyTorch Implementation of Unsupervised Learning of Scene Flow Estimation Fusing with Local Rigidity

UnRigidFlow This is the official PyTorch implementation of UnRigidFlow (IJCAI2019). Here are two sample results (~10MB gif for each) of our unsupervis

Liang Liu 28 Nov 16, 2022
Athena is the only tool that you will ever need to optimize your portfolio.

Athena Portfolio optimization is the process of selecting the best portfolio (asset distribution), out of the set of all portfolios being considered,

Indrajit 1 Mar 25, 2022
A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains (IJCV submission)

wsss-analysis The code of: A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains, arXiv pre-print 2019 paper.

Lyndon Chan 48 Dec 18, 2022
Joint Detection and Identification Feature Learning for Person Search

Person Search Project This repository hosts the code for our paper Joint Detection and Identification Feature Learning for Person Search. The code is

712 Dec 17, 2022
This is the repository for paper NEEDLE: Towards Non-invertible Backdoor Attack to Deep Learning Models.

This is the repository for paper NEEDLE: Towards Non-invertible Backdoor Attack to Deep Learning Models.

1 Oct 25, 2021
Neural-net-from-scratch - A simple Neural Network from scratch in Python using the Pymathrix library

A Simple Neural Network from scratch A Simple Neural Network from scratch in Pyt

Youssef Chafiqui 2 Jan 07, 2022
EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction

EquiBind: geometric deep learning for fast predictions of the 3D structure in which a small molecule binds to a protein

Hannes Stärk 355 Jan 03, 2023
Offical code for the paper: "Growing 3D Artefacts and Functional Machines with Neural Cellular Automata" https://arxiv.org/abs/2103.08737

Growing 3D Artefacts and Functional Machines with Neural Cellular Automata Video of more results: https://www.youtube.com/watch?v=-EzztzKoPeo Requirem

Robotics Evolution and Art Lab 51 Jan 01, 2023
Implementation of ICCV 2021 oral paper -- A Novel Self-Supervised Learning for Gaussian Mixture Model

SS-GMM Implementation of ICCV 2021 oral paper -- Self-Supervised Image Prior Learning with GMM from a Single Noisy Image with supplementary material R

HUST-The Tan Lab 4 Dec 05, 2022
Pytorch Lightning code guideline for conferences

Deep learning project seed Use this seed to start new deep learning / ML projects. Built in setup.py Built in requirements Examples with MNIST Badges

Pytorch Lightning 1k Jan 02, 2023
TorchMultimodal is a PyTorch library for training state-of-the-art multimodal multi-task models at scale.

TorchMultimodal (Alpha Release) Introduction TorchMultimodal is a PyTorch library for training state-of-the-art multimodal multi-task models at scale.

Meta Research 663 Jan 06, 2023
DCSAU-Net: A Deeper and More Compact Split-Attention U-Net for Medical Image Segmentation

DCSAU-Net: A Deeper and More Compact Split-Attention U-Net for Medical Image Segmentation By Qing Xu, Wenting Duan and Na He Requirements pytorch==1.1

Qing Xu 20 Dec 09, 2022
基于Pytorch实现优秀的自然图像分割框架!(包括FCN、U-Net和Deeplab)

语义分割学习实验-基于VOC数据集 usage: 下载VOC数据集,将JPEGImages SegmentationClass两个文件夹放入到data文件夹下。 终端切换到目标目录,运行python train.py -h查看训练 (torch) Li Xiang 28 Dec 21, 2022

This is the official pytorch implementation of Student Helping Teacher: Teacher Evolution via Self-Knowledge Distillation(TESKD)

Student Helping Teacher: Teacher Evolution via Self-Knowledge Distillation (TESKD) By Zheng Li[1,4], Xiang Li[2], Lingfeng Yang[2,4], Jian Yang[2], Zh

Zheng Li 9 Sep 26, 2022
The implementation of FOLD-R++ algorithm

FOLD-R-PP The implementation of FOLD-R++ algorithm. The target of FOLD-R++ algorithm is to learn an answer set program for a classification task. Inst

13 Dec 23, 2022