graphs-10

Molecular Property Prediction

3260 papers • 126 benchmarks • 313 datasets

Molecular property prediction is the task of predicting the properties of a molecule from its structure.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in molecular-property-prediction-43

Trend

Dataset

Best Model

Actions

BBBP

ClinTox

Tox21

Libraries

i

Use these libraries to find molecular-property-prediction-43 models and implementations

PaddlePaddle/PaddleHelix

2 papers 776

Datasets

MoleculeNet

QM9

Tox21

SIDER

AIDS Antiviral Screen

Molecule3D

Subtasks

3D Geometry Prediction NMR J-coupling Odor Descriptor Prediction mixture property prediction

Most implemented papers

Neural Message Passing for Quantum Chemistry

O. Vinyals, J. Gilmer, George E. Dahl, S. Schoenholz, Patrick F. Riley•Mon Apr 03 2017

Using MPNNs, state of the art results on an important molecular property prediction benchmark are demonstrated and it is believed future work should focus on datasets with larger molecules or more accurate ground truth labels.

8540

Content

SIDER

BACE

Lipophilicity

FreeSolv

HIV dataset

QM7

QM8

QM9

ToxCast

ESOL

MoleculeNet

MUV

Clearance

HIV

PCBA

graphcore/ogb-lsc-pcqm4mv2

2 papers 57

Photoswitch

PubChem18

Multi-Labelled SMILES Odors dataset

0

Paper Graph

Strategies for Pre-training Graph Neural Networks

Weihua Hu, M. Zitnik, Bowen Liu, J. Leskovec, Percy Liang, Joseph Gomes, V. Pande•Tue May 28 2019

A new strategy and self-supervised methods for pre-training Graph Neural Networks (GNNs) that avoids negative transfer and improves generalization significantly across downstream tasks, leading up to 9.4% absolute improvements in ROC-AUC over non-pre-trained models and achieving state-of-the-art performance for molecular property prediction and protein function prediction.

1678 0

Paper Graph

InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization

Jordan Hoffmann, Jian Tang, Fan-Yun Sun•Tue Jul 30 2019

Experimental results on the tasks of graph classification and molecular property prediction show that InfoGraph is superior to state-of-the-art baselines and InfoGraph* can achieve performance competitive with state- of- the-art semi-supervised models.

976 0

Paper Graph

Analyzing Learned Molecular Representations for Property Prediction

Connor W. Coley, Kevin Yang, Wengong Jin, T. Jaakkola, R. Barzilay, Kyle Swanson, Philipp Eiden, Hua Gao, A. Guzman-Perez, Timothy Hopper, Brian P. Kelley, M. Mathea, Andrew Palmer, Volker Settels, K. Jensen•Mon Apr 01 2019

A graph convolutional model is introduced that consistently matches or outperforms models using fixed molecular descriptors as well as previous graph neural architectures on both public and proprietary data sets.

1556 0

Paper Graph

Self-Supervised Graph Transformer on Large-Scale Molecular Data

Yatao Bian, Yu Rong, Junzhou Huang, Tingyang Xu, Wei-yang Xie, Ying Wei, Wen-bing Huang•Wed Jun 17 2020

A novel framework, GROVER, which stands for Graph Representation frOm self-supervised mEssage passing tRansformer, which allows it to be trained efficiently on large-scale molecular dataset without requiring any supervision, thus being immunized to the two issues mentioned above.

942 0

Paper Graph

ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction

Bharath Ramsundar, Seyone Chithrananda, Gabriel Grand•Sun Oct 18 2020

This work makes one of the first attempts to systematically evaluate transformers on molecular property prediction tasks via the ChemBERTa model, and suggests that transformers offer a promising avenue of future work for molecular representation learning and property prediction.

605 0

Paper Graph

Molecule3D: A Benchmark for Predicting 3D Geometries from Molecular Graphs

Maho Nakata, Zhao Xu, Youzhi Luo, Xuan Zhang, Xinyi Xu, Yaochen Xie, Meng Liu, Kaleb Dickerson, Cheng Deng, Shuiwang Ji•Wed Sep 29 2021

This work proposes to predict the ground-state 3D geometries from molecular graphs using machine learning methods using density functional theory (DFT), and implements two baseline methods that either predict the pairwise distance between atoms or atom coordinates in 3D space.

47 0

Paper Graph

A Molecular Multimodal Foundation Model Associating Molecule Graphs with Natural Language

Anyi Rao, Ji-rong Wen, Yujie Zhou, Bing Su, Zhiwu Lu, Dazhao Du, Zhao-Qing Yang, Jiangmeng Li, Haoran Sun•Sun Sep 11 2022

A molecular multimodal foundation model which is pretrained from molecular graphs and their semantically related textual data via contrastive learning which enhances molecular property prediction and possesses capability to generate meaningful molecular graphs from natural language descriptions is proposed.

159 0

Paper Graph

Isotropic Gaussian Processes on Finite Spaces of Graphs

Andreas Krause, Viacheslav Borovitskiy, Mohammad Reza Karimi, Vignesh Ram Somnath•Wed Nov 02 2022

An efficient computational technique is proposed for the ostensibly intractable problem of evaluating Gaussian process priors' kernels, making such Gaussian processes usable within the usual toolboxes and downstream applications.

8 0

Paper Graph

Path-Augmented Graph Transformer Network

T. Jaakkola, R. Barzilay, Benson Chen•Tue May 28 2019

This work proposes Path-Augmented Graph Transformer Networks (PAGTN) that are explicitly built on longer-range dependencies in graph-structured data and uses path features in molecular graphs to create global attention layers.

61 0

Paper Graph

Adding a benchmark result helps the community track progress.