Pin Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Pin Chen has a total of 123 co-authors. Here are the top co-authors by number of shared publications:
- Yutong Lu
Sun Yat-sen University
6 shared publications Yutong Lu
National Supercomputer Center in Guangzhou School of Computer Science and Engineering Sun Yat‐sen University Guangzhou China
5 shared publications- Dong-Dong Zhou
MOE Key Laboratory of Bioinorganic and Synthetic Chemistry, GBRCE for Functional Molecular Engineering, School of Chemistry, IGCME, Sun Yat-Sen University, Guangzhou 510275, China
4 shared publications Hui Yan
National Supercomputer Center in Guangzhou, School of Computer Science and Engineering, Sun Yat-Sen University , Guangzhou 510006, China
3 shared publications- Jie-Peng Zhang
MOE Key Laboratory of Bioinorganic and Synthetic Chemistry, GBRCE for Functional Molecular Engineering, School of Chemistry, IGCME, Sun Yat-Sen University, Guangzhou 510275, China
3 shared publications - Dan Huang
School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
2 shared publications
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 9486-9497
UniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 9486-9497
Open AccessUniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials
- Open Access
- 18 November 2025
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 40 (3) , 334-351
HPC-AI coupling methodology for scientific applications
Artificial intelligence (AI) technologies have fundamentally transformed numerical-based high-performance computing (HPC) applications with data-driven approaches and endeavored to address existing challenges, e.g., high computational intensity, in various scientific domains. In this study, we explore the scenarios of coupling HPC and AI (HPC-AI) in the context of emerging scientific applications, presenting a novel methodology that incorporates three coupling patterns: surrogate, directive, and coordinate. Each pattern exemplifies a distinct coupling strategy, an AI-driven prerequisite, and typical HPC-AI ensembles. Through case studies in materials science, we demonstrate the application and effectiveness of these patterns. The study highlights technical challenges, performance improvements, and implementation details, providing insight into promising perspectives of HPC-AI coupling. The proposed coupling patterns are applicable not only to materials science but also to other scientific domains, offering valuable guidance for future HPC-AI ensembles in scientific discovery.
- 18 November 2025
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 40 (3) , 334-351
Open AccessHPC-AI coupling methodology for scientific applications
Artificial intelligence (AI) technologies have fundamentally transformed numerical-based high-performance computing (HPC) applications with data-driven approaches and endeavored to address existing challenges, e.g., high computational intensity, in various scientific domains. In this study, we explore the scenarios of coupling HPC and AI (HPC-AI) in the context of emerging scientific applications, presenting a novel methodology that incorporates three coupling patterns: surrogate, directive, and coordinate. Each pattern exemplifies a distinct coupling strategy, an AI-driven prerequisite, and typical HPC-AI ensembles. Through case studies in materials science, we demonstrate the application and effectiveness of these patterns. The study highlights technical challenges, performance improvements, and implementation details, providing insight into promising perspectives of HPC-AI coupling. The proposed coupling patterns are applicable not only to materials science but also to other scientific domains, offering valuable guidance for future HPC-AI ensembles in scientific discovery.
- 24 July 2025
- Published by Springer Nature
- p. 307-318
No-Data-Driven Crystal Structure Prediction via Model-Free Reinforcement Learning
- 24 July 2025
- Published by Springer Nature
- p. 307-318
No-Data-Driven Crystal Structure Prediction via Model-Free Reinforcement Learning
- 7 April 2025
- Published by Springer Nature in CCF Transactions on High Performance Computing
- Vol. 7 (2) , 85-99
Star-gen: an HPC-AI framework for constructing large-scale computational materials database
- 7 April 2025
- Published by Springer Nature in CCF Transactions on High Performance Computing
- Vol. 7 (2) , 85-99
Star-gen: an HPC-AI framework for constructing large-scale computational materials database
- 26 August 2024
- Published by Springer Nature
- p. 313-328
Efficient Coupling Streaming AI and Ensemble Simulations on HPC Clusters
- 26 August 2024
- Published by Springer Nature
- p. 313-328
Efficient Coupling Streaming AI and Ensemble Simulations on HPC Clusters
- Open Access
- 16 June 2024
- Published by Wiley in Materials Genome Engineering Advances
- Vol. 2 (2)
Bond sensitive graph neural networks for predicting high temperature superconductors
Finding high temperature superconductors (HTS) has been a continuing challenge due to the difficulty in predicting the transition temperature (Tc) of superconductors. Recently, the efficiency of predicting Tc has been greatly improved via machine learning (ML). Unfortunately, prevailing ML models have not shown adequate generalization ability to find new HTS, yet. In this work, a graph neural network model is trained to predict the maximal Tc (Tcmax) of various materials. Our model reveals a close connection between Tcmax and chemical bonds. It suggests that shorter bond lengths are favored by high Tc, which is in coherence with previous domain knowledge. More importantly, it also indicates that chemical bonds consisting of some specific chemical elements are responsible for high Tc, which is new even to the human experts. It can provide a convenient guidance to the materials scientists in search of HTS.
- 16 June 2024
- Published by Wiley in Materials Genome Engineering Advances
- Vol. 2 (2)
Open AccessBond sensitive graph neural networks for predicting high temperature superconductors
Finding high temperature superconductors (HTS) has been a continuing challenge due to the difficulty in predicting the transition temperature (Tc) of superconductors. Recently, the efficiency of predicting Tc has been greatly improved via machine learning (ML). Unfortunately, prevailing ML models have not shown adequate generalization ability to find new HTS, yet. In this work, a graph neural network model is trained to predict the maximal Tc (Tcmax) of various materials. Our model reveals a close connection between Tcmax and chemical bonds. It suggests that shorter bond lengths are favored by high Tc, which is in coherence with previous domain knowledge. More importantly, it also indicates that chemical bonds consisting of some specific chemical elements are responsible for high Tc, which is new even to the human experts. It can provide a convenient guidance to the materials scientists in search of HTS.
- Open Access
Searching High Temperature Superconductors with the assistance of Graph Neural Networks
Predicting high temperature superconductors has long been a great challenge. A major difficulty is how to predict the transition temperature Tc of superconductors. Recently, progress in material informatics has led to a number of machine learning models predicting Tc, which greatly improves the efficiency of prediction. Unfortunately, prevailing models have not shown adequate physical rationality and generalization ability to find new high temperature superconductors, yet. In this work, in order to give a trustable prediction on the unexplored materials, we built a bond-sensitive graph neural network (BSGNN), which is optimized to process the information of chemical bond and electron interaction in the crystal lattice, to predict the Tc maximum of each type of superconducting materials. On the basis of the domain knowledge considered in the data preparation and algorithm design, our model revealed a relevance between the Tc-Tc maximum and chemical bonds. The results indicate that shorter bond length is favored by high Tc, which is in accordance with previous human experience. Moreover, it also shows that some specific chemical elements are favored by high Tc, which is beyond what human experts already knew. It gives a convenient guidance for searching high temperature superconductors in materials database, by ruling out the materials that could never have high Tc.
Open AccessSearching High Temperature Superconductors with the assistance of Graph Neural Networks
Predicting high temperature superconductors has long been a great challenge. A major difficulty is how to predict the transition temperature Tc of superconductors. Recently, progress in material informatics has led to a number of machine learning models predicting Tc, which greatly improves the efficiency of prediction. Unfortunately, prevailing models have not shown adequate physical rationality and generalization ability to find new high temperature superconductors, yet. In this work, in order to give a trustable prediction on the unexplored materials, we built a bond-sensitive graph neural network (BSGNN), which is optimized to process the information of chemical bond and electron interaction in the crystal lattice, to predict the Tc maximum of each type of superconducting materials. On the basis of the domain knowledge considered in the data preparation and algorithm design, our model revealed a relevance between the Tc-Tc maximum and chemical bonds. The results indicate that shorter bond length is favored by high Tc, which is in accordance with previous human experience. Moreover, it also shows that some specific chemical elements are favored by high Tc, which is beyond what human experts already knew. It gives a convenient guidance for searching high temperature superconductors in materials database, by ruling out the materials that could never have high Tc.
- Open Access
- 25 October 2023
- Published by American Chemical Society (ACS) in ChemRxiv
Molecular insights into the formation of drug-polymer inclusion complex
Drug-polymer inclusion complex (IC) has been viewed as a novel solid form of drugs for property modification. Nonetheless, our understanding of the formation mechanism remains limited. This work aims to provide insight into the molecular processes governing the structural construction of carbamazepine (CBZ) and griseofulvin (GSF) channel-type ICs in the presence of guest polymers. Leveraging microdroplet melt crystallization, we successfully unveiled the single-crystal structures of these ICs, enabling theoretical analysis. Our investigation, which encompasses structural analysis, density functional theory calculations, and molecular dynamics (MD) simulations, elucidates the disparity between CBZ and GSF channels in terms of their autonomy in the absence of guest polymers. CBZ molecules can spontaneously assemble into stable channel structures independently, capitalizing on their unique mortise-tenon architecture and robust π...π interactions. In contrast, GSF channels lack sufficient support from weak Cl...O and C-H...π intermolecular interactions and necessitate the insertion of guest molecules to stabilize their structures. Regardless of the structural reliance on guest polymers, channel size is determined by the size, shape, and conformation of the host molecules, as well as intermolecular interactions. Interestingly, while the eleven structurally determined drug-polymer ICs adopt diverse approaches to construct channel structures, their channel sizes consistently fall within a narrow range of 3.86-5.18 Å, slightly larger than the radial diameter of the guest polymers (2.83-3.50 Å). Consequently, we propose that a crucial prerequisite for the formation of drug-polymer ICs is that the host molecules have the capacity to self-assemble into a porous structure with accommodating channels for guest polymers. Additionally, our results confirm the efficacy of microdroplet melt crystallization in rapidly synthesizing drug-polymer ICs and cultivating their single crystals of high quality and sufficient size. This achievement overcomes the challenges associated with structure elucidation and promises to promote further research into the formation mechanism of drug-polymer ICs. We anticipate that these findings will inspire continued exploration of this novel solid form, facilitating theoretical predictions and practical applications in pharmaceutical development.
- 25 October 2023
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessMolecular insights into the formation of drug-polymer inclusion complex
Drug-polymer inclusion complex (IC) has been viewed as a novel solid form of drugs for property modification. Nonetheless, our understanding of the formation mechanism remains limited. This work aims to provide insight into the molecular processes governing the structural construction of carbamazepine (CBZ) and griseofulvin (GSF) channel-type ICs in the presence of guest polymers. Leveraging microdroplet melt crystallization, we successfully unveiled the single-crystal structures of these ICs, enabling theoretical analysis. Our investigation, which encompasses structural analysis, density functional theory calculations, and molecular dynamics (MD) simulations, elucidates the disparity between CBZ and GSF channels in terms of their autonomy in the absence of guest polymers. CBZ molecules can spontaneously assemble into stable channel structures independently, capitalizing on their unique mortise-tenon architecture and robust π...π interactions. In contrast, GSF channels lack sufficient support from weak Cl...O and C-H...π intermolecular interactions and necessitate the insertion of guest molecules to stabilize their structures. Regardless of the structural reliance on guest polymers, channel size is determined by the size, shape, and conformation of the host molecules, as well as intermolecular interactions. Interestingly, while the eleven structurally determined drug-polymer ICs adopt diverse approaches to construct channel structures, their channel sizes consistently fall within a narrow range of 3.86-5.18 Å, slightly larger than the radial diameter of the guest polymers (2.83-3.50 Å). Consequently, we propose that a crucial prerequisite for the formation of drug-polymer ICs is that the host molecules have the capacity to self-assemble into a porous structure with accommodating channels for guest polymers. Additionally, our results confirm the efficacy of microdroplet melt crystallization in rapidly synthesizing drug-polymer ICs and cultivating their single crystals of high quality and sufficient size. This achievement overcomes the challenges associated with structure elucidation and promises to promote further research into the formation mechanism of drug-polymer ICs. We anticipate that these findings will inspire continued exploration of this novel solid form, facilitating theoretical predictions and practical applications in pharmaceutical development.
- Open Access
- 16 June 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (24) , eadg2229
Heat-driven molecule gatekeepers in MOF membrane for record-high H 2 selectivity
Hydrogen/carbon dioxide (H 2 /CO 2 ) separation for sustainable energy is in desperate need of reliable membranes at high temperatures. Molecular sieve membranes take their nanopores to differentiate sizes between H 2 and CO 2 but have compromised at a marked loss of selectivity at high temperatures owing to diffusion activation of CO 2 . We used molecule gatekeepers that were locked in the cavities of the metal-organic framework membrane to meet this challenge. Ab initio calculations and in situ characterizations demonstrate that the molecule gatekeepers make a notable move at high temperatures to dynamically reshape the sieving apertures as being extremely tight for CO 2 and restitute with cool conditions. The H 2 /CO 2 selectivity was improved by an order of magnitude at 513 kelvin (K) relative to that at the ambient temperature.
- 16 June 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (24) , eadg2229
Open AccessHeat-driven molecule gatekeepers in MOF membrane for record-high H 2 selectivity
Hydrogen/carbon dioxide (H 2 /CO 2 ) separation for sustainable energy is in desperate need of reliable membranes at high temperatures. Molecular sieve membranes take their nanopores to differentiate sizes between H 2 and CO 2 but have compromised at a marked loss of selectivity at high temperatures owing to diffusion activation of CO 2 . We used molecule gatekeepers that were locked in the cavities of the metal-organic framework membrane to meet this challenge. Ab initio calculations and in situ characterizations demonstrate that the molecule gatekeepers make a notable move at high temperatures to dynamically reshape the sieving apertures as being extremely tight for CO 2 and restitute with cool conditions. The H 2 /CO 2 selectivity was improved by an order of magnitude at 513 kelvin (K) relative to that at the ambient temperature.
- 15 May 2023
- Published by Elsevier in International Journal of Biological Macromolecules
- Vol. 243, 124887
Preparation and physicochemical properties characterization of hesperetin-grafted pectin conjugate
- 15 May 2023
- Published by Elsevier in International Journal of Biological Macromolecules
- Vol. 243, 124887
Preparation and physicochemical properties characterization of hesperetin-grafted pectin conjugate