Bidirectional Generation of Structure and Properties Through a Single Molecular Foundation Model

19 Nov 2022  ·  Jinho Chang, Jong Chul Ye ·

The recent success of large foundation models in artificial intelligence has prompted the emergence of chemical pre-trained models. Despite the growing interest in large molecular pre-trained models that provide informative representations for downstream tasks, attempts for multimodal pre-training approaches on the molecule domain were limited. To address this, we present a novel multimodal molecular pre-trained model that incorporates the modalities of structure and biochemical properties, drawing inspiration from recent advances in multimodal learning techniques. Our proposed model pipeline of data handling and training objectives aligns the structure/property features in a common embedding space, which enables the model to regard bidirectional information between the molecules' structure and properties. These contributions emerge synergistic knowledge, allowing us to tackle both multimodal and unimodal downstream tasks through a single model. Through extensive experiments, we demonstrate that our model shows remarkable capabilities in solving various meaningful chemical challenges, including conditional molecule generation, property prediction, molecule classification, and reaction prediction.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Molecular Property Prediction BACE SPMM ROC-AUC 83.0 # 5
RMSE 1.108 # 1
Molecular Property Prediction BBBP SPMM ROC-AUC 73.3 # 2
Molecular Property Prediction Clearance SPMM RMSE 44.752 # 1
Molecular Property Prediction ClinTox SPMM ROC-AUC 91.0 # 3
Molecular Property Prediction ESOL SPMM RMSE 0.810 # 3
Molecular Property Prediction FreeSolv SPMM RMSE 1.859 # 2
Molecular Property Prediction Lipophilicity SPMM RMSE 0.706 # 4
Molecular Property Prediction SIDER SPMM ROC-AUC 64.7 # 8


No methods listed for this paper. Add relevant methods here