Gabriel A. Pinheiro
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Gabriel A. Pinheiro has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Juarez L. F. Da Silva
University of São Paulo Av. Trabalhador São-Carlense 400
4 shared publications - Marcos Gonçalves Quiles
Federal University of São Paulo
4 shared publications - Alex Silva de Moraes
Department of Chemistry, Universidade Federal de São Carlos, São Carlos 13565-905, SP, Brazil
1 shared publication - Luis Gustavo Dias
University of São Paulo
1 shared publication - Mauro Chierici Lopes
Department of Chemistry
1 shared publication - Piero A. L. Ribeiro
Institute of Science and Technology, Federal University of São Paulo (Unifesp)
1 shared publication
Latest Publications
- 29 September 2025
- Published by Springer Nature
- p. 426-442
Mitigating Data Scarcity in Polymer Property Prediction via Multi-task Auxiliary Learning
- 29 September 2025
- Published by Springer Nature
- p. 426-442
Mitigating Data Scarcity in Polymer Property Prediction via Multi-task Auxiliary Learning
- 1 August 2024
- Published by Springer Nature
- p. 387-401
Enhancing Low-Cost Molecular Property Prediction with Contrastive Learning on SMILES Representations
- 1 August 2024
- Published by Springer Nature
- p. 387-401
Enhancing Low-Cost Molecular Property Prediction with Contrastive Learning on SMILES Representations
- 19 September 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 62 (19) , 4702-4712
Screening of the Role of the Chemical Structure in the Electrochemical Stability Window of Ionic Liquids: DFT Calculations Combined with Data Mining
Ionic liquids have attracted the attention of researchers as possible electrolytes for electrochemical energy storage devices. However, their properties, such as the electrochemical stability window (ESW), ionic conductivity, and diffusivity, are influenced both by the chemical structures of cations and anions and by their combinations. Most studies in the literature focus on the understanding of common ionic liquids, and little effort has been made to find ways to improve our atomistic understanding of those systems. The goal of this paper is to explore the structural characteristics of cations and anions that form ionic liquids that can expand the HOMO/LUMO gap, a property directly linked to the ESW of the electrolyte. For that, we design a framework for randomly generating new ions by combining their fragments. Within this framework, we generate about 104 cations and 104 anions and fully optimize their structures using density functional theory. Our calculations show that aromatic cations are less stable ionic liquids than aliphatic ones, an expected result if chemical rationale is used. More importantly, we can improve the gap by adding electron-donating and electron-withdrawing functional groups to the cations and anions, respectively. The increase can be about 2 V, depending on the case. This improvement is reflected in a wider ESW.
- 19 September 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 62 (19) , 4702-4712
Screening of the Role of the Chemical Structure in the Electrochemical Stability Window of Ionic Liquids: DFT Calculations Combined with Data Mining
Ionic liquids have attracted the attention of researchers as possible electrolytes for electrochemical energy storage devices. However, their properties, such as the electrochemical stability window (ESW), ionic conductivity, and diffusivity, are influenced both by the chemical structures of cations and anions and by their combinations. Most studies in the literature focus on the understanding of common ionic liquids, and little effort has been made to find ways to improve our atomistic understanding of those systems. The goal of this paper is to explore the structural characteristics of cations and anions that form ionic liquids that can expand the HOMO/LUMO gap, a property directly linked to the ESW of the electrolyte. For that, we design a framework for randomly generating new ions by combining their fragments. Within this framework, we generate about 104 cations and 104 anions and fully optimize their structures using density functional theory. Our calculations show that aromatic cations are less stable ionic liquids than aliphatic ones, an expected result if chemical rationale is used. More importantly, we can improve the gap by adding electron-donating and electron-withdrawing functional groups to the cations and anions, respectively. The increase can be about 2 V, depending on the case. This improvement is reflected in a wider ESW.
- 31 August 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 62 (17) , 3948-3960
SMICLR: Contrastive Learning on Multiple Molecular Representations for Semisupervised and Unsupervised Representation Learning
Machine learning as a tool for chemical space exploration broadens horizons to work with known and unknown molecules. At its core lies molecular representation, an essential key to improve learning about structure-property relationships. Recently, contrastive frameworks have been showing impressive results for representation learning in diverse domains. Therefore, this paper proposes a contrastive framework that embraces multimodal molecular data. Specifically, our approach jointly trains a graph encoder and an encoder for the simplified molecular-input line-entry system (SMILES) string to perform the contrastive learning objective. Since SMILES is the basis of our method, i.e., we built the molecular graph from the SMILES, we call our framework as SMILES Contrastive Learning (SMICLR). When stacking a nonlinear regressor on the SMICLR's pretrained encoder and fine-tuning the entire model, we reduced the prediction error by, on average, 44% and 25% for the energetic and electronic properties of the QM9 data set, respectively, over the supervised baseline. We further improved our framework's performance when applying data augmentations in each molecular-input representation. Moreover, SMICLR demonstrated competitive representation learning results in an unsupervised setting.
- 31 August 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 62 (17) , 3948-3960
SMICLR: Contrastive Learning on Multiple Molecular Representations for Semisupervised and Unsupervised Representation Learning
Machine learning as a tool for chemical space exploration broadens horizons to work with known and unknown molecules. At its core lies molecular representation, an essential key to improve learning about structure-property relationships. Recently, contrastive frameworks have been showing impressive results for representation learning in diverse domains. Therefore, this paper proposes a contrastive framework that embraces multimodal molecular data. Specifically, our approach jointly trains a graph encoder and an encoder for the simplified molecular-input line-entry system (SMILES) string to perform the contrastive learning objective. Since SMILES is the basis of our method, i.e., we built the molecular graph from the SMILES, we call our framework as SMILES Contrastive Learning (SMICLR). When stacking a nonlinear regressor on the SMICLR's pretrained encoder and fine-tuning the entire model, we reduced the prediction error by, on average, 44% and 25% for the energetic and electronic properties of the QM9 data set, respectively, over the supervised baseline. We further improved our framework's performance when applying data augmentations in each molecular-input representation. Moreover, SMICLR demonstrated competitive representation learning results in an unsupervised setting.