Drug Discovery

249 papers with code • 27 benchmarks • 20 datasets

Drug discovery is the task of applying machine learning to discover new candidate drugs.

( Image credit: A Turing Test for Molecular Generators )

Benchmarks

Add a Result

These leaderboards are used to track progress in Drug Discovery

Dataset	Best Model	Compare
Tox21	SSVAE with multiple SMILES	See all
QM9	MXMNet	See all
MUV	TrimNet	See all
HIV dataset	GraphConv + dummy super node + focal loss	See all
ToxCast	TrimNet	See all
BACE	TrimNet	See all
BindingDB	AttentionSiteDTI	See all
LIT-PCBA(MAPK1)	GLAM	See all
LIT-PCBA(ALDH1)	GLAM	See all
LIT-PCBA(ESR1_ant)	GLAM	See all
LIT-PCBA(KAT2A)	GLAM	See all
PCBA	GraphConv + dummy super node	See all
SIDER	Ensemble locally constant networks	See all
ClinTox	TrimNet	See all
BBBP	ProtoW-L2	See all
egfr-inh	Multi-input Neural network with Attention	See all
QED	HierG2G	See all
DRD2	HierG2G	See all
PDBbind	Ensemble locally constant networks	See all
ESOL(scaffold)	GLAM	See all
Lipophilicity(scaffold)	GLAM	See all
FreeSolv(scaffold)	GLAM	See all
BACE(scaffold)	GLAM	See all
BBBP(scaffold)	GLAM	See all
SIDER(scaffold)	GLAM	See all
Tox21(scaffold)	GLAM	See all
ToxCast(scaffold)	GLAM	See all

Show all 27 benchmarks

Collapse benchmarks

Libraries

Use these libraries to find Drug Discovery models and implementations

kexinhuang12345/DeepPurpose

2 papers

708

PaddlePaddle/PaddleHelix

2 papers

658

gnina/gnina

2 papers

394

mattragoza/liGAN

2 papers

170

Datasets

Subtasks

Most implemented papers

Most implemented Social Latest No code

Semi-Supervised Classification with Graph Convolutional Networks

dmlc/dgl • • 9 Sep 2016

We present a scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs.

Paper
Code

Neural Message Passing for Quantum Chemistry

brain-research/mpnn • • ICML 2017

Supervised learning on molecules has incredible potential to be useful in chemistry, drug discovery, and materials science.

Paper
Code

Gated Graph Sequence Neural Networks

dmlc/dgl • • 17 Nov 2015

Graph-structured data appears frequently in domains including chemistry, natural language semantics, social networks, and knowledge bases.

Paper
Code

Junction Tree Variational Autoencoder for Molecular Graph Generation

wengong-jin/icml18-jtnn • • ICML 2018

We evaluate our model on multiple tasks ranging from molecular generation to optimization.

Paper
Code

Self-Normalizing Neural Networks

bioinf-jku/SNNs • • NeurIPS 2017

We introduce self-normalizing neural networks (SNNs) to enable high-level abstract representations.

Paper
Code

Convolutional Networks on Graphs for Learning Molecular Fingerprints

HIPS/neural-fingerprint • • NeurIPS 2015

We introduce a convolutional neural network that operates directly on graphs.

Paper
Code

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges

MeuwlyGroup/PhysNet • • J. Chem. Theory Comput. 2019

Further, two new datasets are generated in order to probe the performance of ML models for describing chemical reactions, long-range interactions, and condensed phase systems.

Paper
Code

Molecule Attention Transformer

gmum/MAT • • 19 Feb 2020

Designing a single neural network architecture that performs competitively across a range of molecule property prediction tasks remains largely an open challenge, and its solution may unlock a widespread use of deep learning in the drug discovery industry.

Paper
Code