Home

Deep Learning Meets Chemistry Home ·

Code

sPhysNet-MT⁴: GitHub models_cal models_exp

sPhysNet and PhysNet²: GitHub

Extended Functional Groups²: GitHub

Frag20_prepare²: GitHub

DTNN_7ib¹: GitHub Tutorial

A3D-PNAConv³: GitHub

CSTShift⁵: GitHub trained_models

Datasets

Name

Size

Download

Frag20-solv-678k & FreeSolv-PHYSPROP-14k⁴	678,916 / 14,339	Download
Frag20²	566296	Download
Plati20²	20972	Download
CSD20²	39816	Download
QM9_M¹	133885	Download
eMol9_C_M¹	88234(9959)	Download
Plati_C_M¹	4076(74)	Download
Property¹		Download
Frag20-Sol-100k³	100000	Download
NMRShiftDB2_DFT⁵	26353 C, 11983 H	Download

Reference

1. Predicting Molecular Energy Using Force-Field Optimized Geometries and Atomic Vector Representations Learned from an Improved Deep Tensor Neural Network

2. Dataset Construction to Explore Chemical Space with 3D Geometry and Deep Learning

3. Accurate prediction of aqueous solvation free energies using 3D atomic feature-based graph neural network with transfer learning

4. Multi-task Deep Ensemble Prediction of Molecular Energetics in Solution: From Quantum Mechanics to Experimental Properties

5. Accurate Prediction of NMR Chemical Shifts: Integrating DFT calculations with 3D Graph Neural Networks

Yingkai Zhang's Lab @ NYU