Structure to Property: Chemical Element Embeddings and a Deep Learning Approach for Accurate Prediction of Chemical Properties

Shokirbek Shermukhamedov, Dilorom Mamurjonova, Michael Probst
9/17/2023
physics.chem-phcond-mat.mtrl-scics.LG

Official code: dmamur/elembert (22 stars, last commit 2025-05-09).

Abstract

We introduce the elEmBERT model for chemical classification tasks. It is based on deep learning techniques, such as a multilayer encoder architecture. We demonstrate the opportunities offered by our approach on sets of organic, inorganic and crystalline compounds. In particular, we developed and tested the model using the Matbench and Moleculenet benchmarks, which include crystal properties and drug design-related benchmarks. We also conduct an analysis of vector representations of chemical compounds, shedding light on the underlying patterns in structural data. Our model exhibits exceptional predictive capabilities and proves universally applicable to molecular and material datasets. For instance, on the Tox21 dataset, we achieved an average precision of 96%, surpassing the previously best result by 10%.

View on arXivView PDF

Code Implementations(1)

dmamur/elembertOfficial
225Jupyter NotebookMay 12, 20231 years ago

Similar Papers

Geometric Deep Learning for Molecular Crystal Structure Prediction

85% match2023

Functional Groups are All you Need for Chemically Interpretable Molecular Property Prediction

85% match2025

CrysAtom: Distributed Representation of Atoms for Crystal Property Prediction

84% match1 repo12024

Foundation Models for Discovery and Exploration in Chemical Space

84% match2025

Leveraging Chemistry Foundation Models to Facilitate Structure Focused Retrieval Augmented Generation in Multi-Agent Workflows for Catalyst and Materials Design

84% match2024

Advancements in Molecular Property Prediction: A Survey of Single and Multimodal Approaches

83% match2024

Cite this paper

@article{shermukhamedov2023structure,
  title  = {Structure to Property: Chemical Element Embeddings and a Deep Learning Approach for Accurate Prediction of Chemical Properties},
  author = {Shokirbek Shermukhamedov and Dilorom Mamurjonova and Michael Probst},
  year   = {2023},
  eprint = {2309.09355},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2309.09355v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2309.09355.svg)](https://codeofpaper.com/paper/2309.09355)

Discussion