Chunyang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chunyang Wang has a total of 63 co-authors. Here are the top co-authors by number of shared publications:
- Jianjun Zhao
Key Laboratory of Geographical Processes and Ecological Security in Changbai Mountains, Ministry of Education, School of Geographical Sciences, Northeast Normal University, Changchun, China
2 shared publications - Lisong Ma
State Key Laboratory of North China Crop Improvement and Regulation, College of Horticulture Hebei Agricultural University Baoding China
2 shared publications - Xizhe Sun
State Key Laboratory of North China Crop Improvement and Regulation Hebei Agricultural University Baoding China
2 shared publications - Aiwu Zhou
Department of Pathophysiology, Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, SJTU-SM, Shanghai, China
1 shared publication - Alison Gail Smith
Department of Plant Sciences, University of Cambridge, Cambridge CB2 3EA, UK
1 shared publication Boyang Ke
Hebei Agricultural University
1 shared publication
Latest Publications
- 6 October 2026
- Published by Oxford University Press (OUP) in Plant Physiology
- Vol. 202 (2)
AraVirusPPI: a protein language model-powered method for predicting Arabidopsis thaliana –virus protein–protein interactions
Plant–virus protein–protein interactions (PPIs) play crucial roles in viral infection and host immune responses. However, large-scale experimental characterization of plant–virus PPIs is costly and labor-intensive, suggesting the need to develop computational methods. Recent advances in biological sequence language models are offering unprecedented opportunities to develop machine learning-based plant virus PPI predictors. In this study, we collected publicly available PPI data between Arabidopsis thaliana (A. thaliana) and viruses, establishing a robust dataset for method development. Based on this dataset, we developed AraVirusPPI, a machine learning framework for predicting A. thaliana virus PPIs. AraVirusPPI employs the protein language model ESM Cambrian (ESMC) to encode sequence features and combines these representations with Extreme Gradient Boosting (XGBoost) to build the prediction model. Furthermore, we systematically evaluated multiple protein and genomic language models combined with various classification algorithms, thereby justifying the optimal model architecture adopted in AraVirusPPI. We applied AraVirusPPI to predict PPIs in the A. thaliana–Cabbage leaf curl virus (CabLCV) system. Predicted viral target proteins were compared with experimental datasets from A. thaliana–Turnip mosaic virus (TuMV) and A. thaliana–Pseudomonas syringae (Psy) systems. Network topology and functional enrichment analyses showed that the predicted CabLCV targets in A. thaliana were more similar to TuMV than Psy targets. These results suggest potential shared and distinct host-targeting patterns among pathogens and provide computational evidence supporting the biological significance of the proposed AraVirusPPI predictor.
- 6 October 2026
- Published by Oxford University Press (OUP) in Plant Physiology
- Vol. 202 (2)
AraVirusPPI: a protein language model-powered method for predicting Arabidopsis thaliana –virus protein–protein interactions
Plant–virus protein–protein interactions (PPIs) play crucial roles in viral infection and host immune responses. However, large-scale experimental characterization of plant–virus PPIs is costly and labor-intensive, suggesting the need to develop computational methods. Recent advances in biological sequence language models are offering unprecedented opportunities to develop machine learning-based plant virus PPI predictors. In this study, we collected publicly available PPI data between Arabidopsis thaliana (A. thaliana) and viruses, establishing a robust dataset for method development. Based on this dataset, we developed AraVirusPPI, a machine learning framework for predicting A. thaliana virus PPIs. AraVirusPPI employs the protein language model ESM Cambrian (ESMC) to encode sequence features and combines these representations with Extreme Gradient Boosting (XGBoost) to build the prediction model. Furthermore, we systematically evaluated multiple protein and genomic language models combined with various classification algorithms, thereby justifying the optimal model architecture adopted in AraVirusPPI. We applied AraVirusPPI to predict PPIs in the A. thaliana–Cabbage leaf curl virus (CabLCV) system. Predicted viral target proteins were compared with experimental datasets from A. thaliana–Turnip mosaic virus (TuMV) and A. thaliana–Pseudomonas syringae (Psy) systems. Network topology and functional enrichment analyses showed that the predicted CabLCV targets in A. thaliana were more similar to TuMV than Psy targets. These results suggest potential shared and distinct host-targeting patterns among pathogens and provide computational evidence supporting the biological significance of the proposed AraVirusPPI predictor.
- Open Access
- 9 December 2025
- Published by MDPI AG in Preprints.org
Groundbreaking Structural Advances in Key Catalysts for Nitrogen and Carbon Fixation: Past, Present and Future
Photosynthesis, a crucial component of the global carbon cycle, and biological nitrogen fixation (BNF), a key step in the nitrogen cycle, are both requisite for maintaining life and sustainable agriculture on Earth, in which chlorophyll and nitrogenase play central roles, respectively. In chlorophyll biosynthesis, the reduction of protochlorophyllide to chlorophyllide is catalyzed by either the nitrogenase-like, light-independent DPOR (dark-operative protochlorophyllide oxidoreductase) or the SDR-like, light-dependent LPOR (light-driven protochlorophyllide oxidoreductase), utilizing chemical and light energy. Nitrogenase (N2ase) is the only enzyme capable of reducing inert atmospheric nitrogen, yet only a few bacterial species possess the multi-subunit ATP-dependent dinitrogen reductase. Over the past three decades, significant progress has been made in solving the crystal structures of N2ase, DPOR, and LPOR protein complex, providing valuable mechanistic insights into N2ase, as well as the contrasting structures of the multi-subunit DPOR and single-subunit LPOR. We summarize the structural breakthroughs of these key catalysts for nitrogen and carbon fixation. Protein structural similarities often hint at similar functions and evolutionary relationships. With the ongoing development in protein structure annotation assisted by AlphaFold and the identification of protein structure similarities through structural alignment software, the evolutionary relationship between these key enzymes in photosynthesis and nitrogen fixation, as well as the potential co-origin of these processes, has been uncovered. We further summarize how a porphyrin reduction catalyst evolves from a less efficient ATP-driven DPOR to a highly efficient light-driven LPOR. The key structural milestones achieved in N2ase, DPOR, and LPOR research laid the groundwork for future perspectives, particularly in proposing a direction for a potential nitrogen-fixing scenario using an AI-designed LUN and photosynthesis with bacteriochlorophyll using a light-driven chlide reductase (LCOR), leading to the development of more effective forms of biological nitrogen fixation and photosynthesis.
- 9 December 2025
- Published by MDPI AG in Preprints.org
Open AccessGroundbreaking Structural Advances in Key Catalysts for Nitrogen and Carbon Fixation: Past, Present and Future
Photosynthesis, a crucial component of the global carbon cycle, and biological nitrogen fixation (BNF), a key step in the nitrogen cycle, are both requisite for maintaining life and sustainable agriculture on Earth, in which chlorophyll and nitrogenase play central roles, respectively. In chlorophyll biosynthesis, the reduction of protochlorophyllide to chlorophyllide is catalyzed by either the nitrogenase-like, light-independent DPOR (dark-operative protochlorophyllide oxidoreductase) or the SDR-like, light-dependent LPOR (light-driven protochlorophyllide oxidoreductase), utilizing chemical and light energy. Nitrogenase (N2ase) is the only enzyme capable of reducing inert atmospheric nitrogen, yet only a few bacterial species possess the multi-subunit ATP-dependent dinitrogen reductase. Over the past three decades, significant progress has been made in solving the crystal structures of N2ase, DPOR, and LPOR protein complex, providing valuable mechanistic insights into N2ase, as well as the contrasting structures of the multi-subunit DPOR and single-subunit LPOR. We summarize the structural breakthroughs of these key catalysts for nitrogen and carbon fixation. Protein structural similarities often hint at similar functions and evolutionary relationships. With the ongoing development in protein structure annotation assisted by AlphaFold and the identification of protein structure similarities through structural alignment software, the evolutionary relationship between these key enzymes in photosynthesis and nitrogen fixation, as well as the potential co-origin of these processes, has been uncovered. We further summarize how a porphyrin reduction catalyst evolves from a less efficient ATP-driven DPOR to a highly efficient light-driven LPOR. The key structural milestones achieved in N2ase, DPOR, and LPOR research laid the groundwork for future perspectives, particularly in proposing a direction for a potential nitrogen-fixing scenario using an AI-designed LUN and photosynthesis with bacteriochlorophyll using a light-driven chlide reductase (LCOR), leading to the development of more effective forms of biological nitrogen fixation and photosynthesis.
- Open Access
The Diversity of NLRs in Brassica rapa Pan-genome
The species Brassica rapa is mainly cultivated as a vegetable crop with high economic value. But infectious diseases, such as soft rot disease and clubroot disease can cause severe yield losses. Introducing resistance genes through breeding is an efficient and sustainable way to reduce the susceptibility and improve the yield of crops. Nucleotide-Binding Leucine-Rich Repeat (NLR) genes are the main types of resistance genes. They can recognize pathogen effectors and initiate downstream immune response. However, the intraspecific diversity of NLR genes in Brassica rapa has remained unknown. Based on domain similarity search and machine learning pipeline, we identified 2188 NLRs in 17 accessions of Brassica rapa genomes, which constitute the species-wide pan-NLRome in Brassica rapa . The diversity of the four types of NLR genes are significantly different in the aspects of chromosome location, number stability, integrated domains (IDs), evolutionary trajectory, and positive selection sites. Phylogenetic analyses show TNL-type NLRs whose N terminus contain Toll/interleukin-1 receptor (TIR) domain experienced accession-specific expansion. Moreover, the expanded TNL-type NLRs carry more positively selected amino acid residues which are mainly located on the protein surface based on their 3D structures. These evidences might imply their important biological function in perceiving pathogen effectors. Taken together, our study provides better insights into the diversity and variation of NLR genes in Brassica rapa pangemone, which can facilitate the biofunction assigning, cloning, and subsequent application in breeding programs.
Open AccessThe Diversity of NLRs in Brassica rapa Pan-genome
The species Brassica rapa is mainly cultivated as a vegetable crop with high economic value. But infectious diseases, such as soft rot disease and clubroot disease can cause severe yield losses. Introducing resistance genes through breeding is an efficient and sustainable way to reduce the susceptibility and improve the yield of crops. Nucleotide-Binding Leucine-Rich Repeat (NLR) genes are the main types of resistance genes. They can recognize pathogen effectors and initiate downstream immune response. However, the intraspecific diversity of NLR genes in Brassica rapa has remained unknown. Based on domain similarity search and machine learning pipeline, we identified 2188 NLRs in 17 accessions of Brassica rapa genomes, which constitute the species-wide pan-NLRome in Brassica rapa . The diversity of the four types of NLR genes are significantly different in the aspects of chromosome location, number stability, integrated domains (IDs), evolutionary trajectory, and positive selection sites. Phylogenetic analyses show TNL-type NLRs whose N terminus contain Toll/interleukin-1 receptor (TIR) domain experienced accession-specific expansion. Moreover, the expanded TNL-type NLRs carry more positively selected amino acid residues which are mainly located on the protein surface based on their 3D structures. These evidences might imply their important biological function in perceiving pathogen effectors. Taken together, our study provides better insights into the diversity and variation of NLR genes in Brassica rapa pangemone, which can facilitate the biofunction assigning, cloning, and subsequent application in breeding programs.
- 4 February 2023
- Published by Elsevier in Molecular Plant
- Vol. 16 (3) , 511-513
On the origins and evolution of phytohormone signaling and biosynthesis in plants
- 4 February 2023
- Published by Elsevier in Molecular Plant
- Vol. 16 (3) , 511-513
On the origins and evolution of phytohormone signaling and biosynthesis in plants
- Open Access
- 20 April 2022
- Published by Frontiers Media SA in Frontiers in Genetics
- Vol. 13, 842387
PolyReco: A Method to Automatically Label Collinear Regions and Recognize Polyploidy Events Based on the KS Dotplot
Polyploidization plays a critical role in producing new gene functions and promoting species evolution. Effective identification of polyploid types can be helpful in exploring the evolutionary mechanism. However, current methods for detecting polyploid types have some major limitations, such as being time-consuming and strong subjectivity, etc. In order to objectively and scientifically recognize collinearity fragments and polyploid types, we developed PolyReco method, which can automatically label collinear regions and recognize polyploidy events based on the KS dotplot. Combining with whole-genome collinearity analysis, PolyReco uses DBSCAN clustering method to cluster KS dots. According to the distance information in the x-axis and y-axis directions between the categories, the clustering results are merged based on certain rules to obtain the collinear regions, automatically recognize and label collinear fragments. According to the information of the labeled collinear regions on the y-axis, the polyploidization recognition algorithm is used to exhaustively combine and obtain the genetic collinearity evaluation index of each combination, and then draw the genetic collinearity evaluation index graph. Based on the inflection point on the graph, polyploid types and related chromosomes with polyploidy signal can be detected. The validation experiments showed that the conclusions of PolyReco were consistent with the previous study, which verified the effectiveness of this method. It is expected that this approach can become a reference architecture for other polyploid types classification methods.
- 20 April 2022
- Published by Frontiers Media SA in Frontiers in Genetics
- Vol. 13, 842387
Open AccessPolyReco: A Method to Automatically Label Collinear Regions and Recognize Polyploidy Events Based on the KS Dotplot
Polyploidization plays a critical role in producing new gene functions and promoting species evolution. Effective identification of polyploid types can be helpful in exploring the evolutionary mechanism. However, current methods for detecting polyploid types have some major limitations, such as being time-consuming and strong subjectivity, etc. In order to objectively and scientifically recognize collinearity fragments and polyploid types, we developed PolyReco method, which can automatically label collinear regions and recognize polyploidy events based on the KS dotplot. Combining with whole-genome collinearity analysis, PolyReco uses DBSCAN clustering method to cluster KS dots. According to the distance information in the x-axis and y-axis directions between the categories, the clustering results are merged based on certain rules to obtain the collinear regions, automatically recognize and label collinear fragments. According to the information of the labeled collinear regions on the y-axis, the polyploidization recognition algorithm is used to exhaustively combine and obtain the genetic collinearity evaluation index of each combination, and then draw the genetic collinearity evaluation index graph. Based on the inflection point on the graph, polyploid types and related chromosomes with polyploidy signal can be detected. The validation experiments showed that the conclusions of PolyReco were consistent with the previous study, which verified the effectiveness of this method. It is expected that this approach can become a reference architecture for other polyploid types classification methods.