Xiangning 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
Xiangning Chen has a total of 166 co-authors. Here are the top co-authors by number of shared publications:
- Jingchun Chen
Nevada Institute of Personalized Medicine, University of Nevada, Las Vegas, Las Vegas, NV, USA
13 shared publications - Zhongming Zhao
The University of Texas Health Science Center at Houston
10 shared publications - Peilin Jia
CAS Key Laboratory of Genomic and Precision Medicine, Beijing Institute of Genomics, Chinese Academy of Sciences and China National Center for Bioinformation, Beijing, China
6 shared publications Zhongming Zhao
Center for Precision Health, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA
6 shared publications- Kenneth S. Kendler
Virginia Institute for Psychiatric and Behavioral Genetics, Department of Psychiatry Virginia Commonwealth University Richmond Virginia USA
5 shared publications Travis Mize
Department of Ecology and Evolutionary Biology, University of Colorado Boulder, Boulder, CO, 80309, USA
5 shared publications
Latest Publications
- Open Access
- 5 February 2025
- Published by Springer Nature in Schizophrenia
- Vol. 11 (1) , 1-11
Classification of schizophrenia, bipolar disorder and major depressive disorder with comorbid traits and deep learning algorithms
Many psychiatric disorders share genetic liabilities, but whether these shared liabilities can be utilized to classify and differentiate psychiatric disorders remains unclear. In this study, we use polygenic risk scores (PRSs) of 42 traits comorbid with schizophrenia (SCZ), bipolar disorder (BIP), and major depressive disorder (MDD) to evaluate their utilities. We found that combining target specific PRS with PRSs of comorbid traits can improve the classification of the target disorders. Importantly, without inclusion of PRSs from targeted disorders, we can still classify SCZ (accuracy 0.710 ± 0.008, AUC 0.789 ± 0.011), BIP (accuracy 0.782 ± 0.006, AUC 0.852 ± 0.004), and MDD (accuracy 0.753 ± 0.019, AUC 0.822 ± 0.010). Furthermore, PRSs from comorbid traits alone can effectively differentiate unaffected controls and patients with SCZ, BIP, and MDD (accuracy 0.861 ± 0.003, AUC 0.961 ± 0.041). Our results demonstrate that shared liabilities can be used effectively to improve the classification and differentiation of these disorders. The finding that PRSs from comorbid traits alone can classify and differentiate SCZ, BIP and MDD reasonably well implies that a majority of the risk variants composing target PRSs are shared with comorbid traits. Overall, our results suggest that a data-driven approach may be feasible to classify and differentiate these disorders.
- 5 February 2025
- Published by Springer Nature in Schizophrenia
- Vol. 11 (1) , 1-11
Open AccessClassification of schizophrenia, bipolar disorder and major depressive disorder with comorbid traits and deep learning algorithms
Many psychiatric disorders share genetic liabilities, but whether these shared liabilities can be utilized to classify and differentiate psychiatric disorders remains unclear. In this study, we use polygenic risk scores (PRSs) of 42 traits comorbid with schizophrenia (SCZ), bipolar disorder (BIP), and major depressive disorder (MDD) to evaluate their utilities. We found that combining target specific PRS with PRSs of comorbid traits can improve the classification of the target disorders. Importantly, without inclusion of PRSs from targeted disorders, we can still classify SCZ (accuracy 0.710 ± 0.008, AUC 0.789 ± 0.011), BIP (accuracy 0.782 ± 0.006, AUC 0.852 ± 0.004), and MDD (accuracy 0.753 ± 0.019, AUC 0.822 ± 0.010). Furthermore, PRSs from comorbid traits alone can effectively differentiate unaffected controls and patients with SCZ, BIP, and MDD (accuracy 0.861 ± 0.003, AUC 0.961 ± 0.041). Our results demonstrate that shared liabilities can be used effectively to improve the classification and differentiation of these disorders. The finding that PRSs from comorbid traits alone can classify and differentiate SCZ, BIP and MDD reasonably well implies that a majority of the risk variants composing target PRSs are shared with comorbid traits. Overall, our results suggest that a data-driven approach may be feasible to classify and differentiate these disorders.
- Open Access
- 24 December 2024
- Published by Springer Nature in Research Square
Shared Genetic Architecture Between COVID-19 Severity and Alzheimer's Disease Across European and African Ancestries
The global outbreak of COVID-19, caused by the SARS-CoV-2 virus, has been linked to long-term neurological complications, including an increased risk of Alzheimer's disease (AD) among older adults. However, the precise genetic impact of COVID-19 on long-term AD development remains unclear. This study leveraged genome-wide association study (GWAS) data and genotype data to explore the genetic association between AD and various COVID-19 phenotypes across European ancestry (EA) and African ancestry (AA) cohorts, and the possibility of a causal effect of COVID-19 on AD. We first calculated polygenic risk scores (PRSs) of three COVID-19 phenotypes in AD cases and controls from both EA and AA populations, then determined the genetic associations between COVID-19 PRSs and AD by logistic regression analyses with or without adjusting for age, sex, and APOE genotypes. Significant positive associations were found between AD diagnosis and COVID-19 PRSs in both populations, with the strongest associations identified in the AA population. However, Mendelian randomization (MR) analyses revealed no evidence of a causal effect of COVID-19 phenotypes on AD liability. We explored this finding further through the analysis of shared genomic regions between the COVID-19 phenotypes and AD and found a region of overlap on chromosome 17 that was highly pleiotropic for traits implicating immune function, psychiatric disorders, and lung function phenotypes. These findings suggest that while COVID-19 and AD share overlapping polygenic contributions involving peripheral genes across multiple traits, they lack a direct connection involving core genes that drive the development of their respective pathologies.
- 24 December 2024
- Published by Springer Nature in Research Square
Open AccessShared Genetic Architecture Between COVID-19 Severity and Alzheimer's Disease Across European and African Ancestries
The global outbreak of COVID-19, caused by the SARS-CoV-2 virus, has been linked to long-term neurological complications, including an increased risk of Alzheimer's disease (AD) among older adults. However, the precise genetic impact of COVID-19 on long-term AD development remains unclear. This study leveraged genome-wide association study (GWAS) data and genotype data to explore the genetic association between AD and various COVID-19 phenotypes across European ancestry (EA) and African ancestry (AA) cohorts, and the possibility of a causal effect of COVID-19 on AD. We first calculated polygenic risk scores (PRSs) of three COVID-19 phenotypes in AD cases and controls from both EA and AA populations, then determined the genetic associations between COVID-19 PRSs and AD by logistic regression analyses with or without adjusting for age, sex, and APOE genotypes. Significant positive associations were found between AD diagnosis and COVID-19 PRSs in both populations, with the strongest associations identified in the AA population. However, Mendelian randomization (MR) analyses revealed no evidence of a causal effect of COVID-19 phenotypes on AD liability. We explored this finding further through the analysis of shared genomic regions between the COVID-19 phenotypes and AD and found a region of overlap on chromosome 17 that was highly pleiotropic for traits implicating immune function, psychiatric disorders, and lung function phenotypes. These findings suggest that while COVID-19 and AD share overlapping polygenic contributions involving peripheral genes across multiple traits, they lack a direct connection involving core genes that drive the development of their respective pathologies.
- Open Access
- 27 May 2024
- Published by Elsevier in Human Genetics and Genomics Advances
- Vol. 5 (3) , 100313
Investigating gene functions and single-cell expression profiles of de novo variants in orofacial clefts
- 27 May 2024
- Published by Elsevier in Human Genetics and Genomics Advances
- Vol. 5 (3) , 100313
Open AccessInvestigating gene functions and single-cell expression profiles of de novo variants in orofacial clefts
- Open Access
- 1 April 2024
- Published by MDPI AG in Epigenomes
- Vol. 8 (2) , 14
Unveiling Gene Interactions in Alzheimer’s Disease by Integrating Genetic and Epigenetic Data with a Network-Based Approach
Alzheimer’s Disease (AD) is a complex disease and the leading cause of dementia in older people. We aimed to uncover aspects of AD’s pathogenesis that may contribute to drug repurposing efforts by integrating DNA methylation and genetic data. Implementing the network-based tool, a dense module search of genome-wide association studies (dmGWAS), we integrated a large-scale GWAS dataset with DNA methylation data to identify gene network modules associated with AD. Our analysis yielded 286 significant gene network modules. Notably, the foremost module included the BIN1 gene, showing the largest GWAS signal, and the GNAS gene, the most significantly hypermethylated. We conducted Web-based Cell-type-Specific Enrichment Analysis (WebCSEA) on genes within the top 10% of dmGWAS modules, highlighting monocyte as the most significant cell type (p < 5 × 10−12). Functional enrichment analysis revealed Gene Ontology Biological Process terms relevant to AD pathology (adjusted p < 0.05). Additionally, drug target enrichment identified five FDA-approved targets (p-value = 0.03) for further research. In summary, dmGWAS integration of genetic and epigenetic signals unveiled new gene interactions related to AD, offering promising avenues for future studies.
- 1 April 2024
- Published by MDPI AG in Epigenomes
- Vol. 8 (2) , 14
Open AccessUnveiling Gene Interactions in Alzheimer’s Disease by Integrating Genetic and Epigenetic Data with a Network-Based Approach
Alzheimer’s Disease (AD) is a complex disease and the leading cause of dementia in older people. We aimed to uncover aspects of AD’s pathogenesis that may contribute to drug repurposing efforts by integrating DNA methylation and genetic data. Implementing the network-based tool, a dense module search of genome-wide association studies (dmGWAS), we integrated a large-scale GWAS dataset with DNA methylation data to identify gene network modules associated with AD. Our analysis yielded 286 significant gene network modules. Notably, the foremost module included the BIN1 gene, showing the largest GWAS signal, and the GNAS gene, the most significantly hypermethylated. We conducted Web-based Cell-type-Specific Enrichment Analysis (WebCSEA) on genes within the top 10% of dmGWAS modules, highlighting monocyte as the most significant cell type (p < 5 × 10−12). Functional enrichment analysis revealed Gene Ontology Biological Process terms relevant to AD pathology (adjusted p < 0.05). Additionally, drug target enrichment identified five FDA-approved targets (p-value = 0.03) for further research. In summary, dmGWAS integration of genetic and epigenetic signals unveiled new gene interactions related to AD, offering promising avenues for future studies.
- Open Access
A Model-agnostic Computational Method for Discovering Gene–Phenotype Relationships and Inferring Gene Networks via in silico Gene Perturbation
Background Deep learning architectures have advanced genotype‒phenotype mappings with precision but often obscure the roles of specific genes and their interactions. Our research introduces a model-agnostic computational methodology, capitalizing on the analytical strengths of deep learning models to serve as biological proxies, enabling interpretation of key gene interactions and their impact on phenotypic outcomes. The objective of this research is to refine the understanding of genetic networks in complex traits by leveraging the nuanced decision-making of advanced models. Results Testing was conducted across several computational models representing varying levels of complexity trained on gene expression datasets for the prediction of the Ki-67 biomarker, which is known for its prognostic value in breast cancer. The methodology is capable of using models as proxies to identify biologically significant genes and to infer relevant gene networks from an entirely data-driven analysis. Notably, the model-derived biomarkers (p-values of 0.013 and 0.003) outperformed the conventional Ki-67 biomarker (0.021) in terms of prognostic efficacy. Moreover, our analysis revealed high congruence between model precision and the biological relevance of the genes and gene relationships identified. Furthermore, we demonstrated that the complexity of the identified gene relationships was consistent with the decision-making intricacy of the model, with complex models capturing greater proportions of complex gene–gene interactions (61.2% and 31.1%) than simpler models (4.6%), reinforcing that the approach effectively captures biologically relevant in-model decision-making processes. Conclusions This methodology offers researchers a powerful tool to examine the decision-making processes within their genotype–phenotype mapping models. It accurately identifies critical genes and their interactions, revealing the biological rationale behind model decisions. It also enables comparisons of decision-making between different models. Furthermore, by discovering in-model critical gene networks, our approach helps bridge the gap between research and clinical applications. It facilitates the translation of complex, model-driven genetic discoveries into actionable clinical insights. This capability is pivotal for advancing personalized medicine, as it leverages the precision of deep learning models to uncover biologically relevant genes and gene networks and opens pathways for discovering new gene biomarker combinations and previously unknown gene interactions.
Open AccessA Model-agnostic Computational Method for Discovering Gene–Phenotype Relationships and Inferring Gene Networks via in silico Gene Perturbation
Background Deep learning architectures have advanced genotype‒phenotype mappings with precision but often obscure the roles of specific genes and their interactions. Our research introduces a model-agnostic computational methodology, capitalizing on the analytical strengths of deep learning models to serve as biological proxies, enabling interpretation of key gene interactions and their impact on phenotypic outcomes. The objective of this research is to refine the understanding of genetic networks in complex traits by leveraging the nuanced decision-making of advanced models. Results Testing was conducted across several computational models representing varying levels of complexity trained on gene expression datasets for the prediction of the Ki-67 biomarker, which is known for its prognostic value in breast cancer. The methodology is capable of using models as proxies to identify biologically significant genes and to infer relevant gene networks from an entirely data-driven analysis. Notably, the model-derived biomarkers (p-values of 0.013 and 0.003) outperformed the conventional Ki-67 biomarker (0.021) in terms of prognostic efficacy. Moreover, our analysis revealed high congruence between model precision and the biological relevance of the genes and gene relationships identified. Furthermore, we demonstrated that the complexity of the identified gene relationships was consistent with the decision-making intricacy of the model, with complex models capturing greater proportions of complex gene–gene interactions (61.2% and 31.1%) than simpler models (4.6%), reinforcing that the approach effectively captures biologically relevant in-model decision-making processes. Conclusions This methodology offers researchers a powerful tool to examine the decision-making processes within their genotype–phenotype mapping models. It accurately identifies critical genes and their interactions, revealing the biological rationale behind model decisions. It also enables comparisons of decision-making between different models. Furthermore, by discovering in-model critical gene networks, our approach helps bridge the gap between research and clinical applications. It facilitates the translation of complex, model-driven genetic discoveries into actionable clinical insights. This capability is pivotal for advancing personalized medicine, as it leverages the precision of deep learning models to uncover biologically relevant genes and gene networks and opens pathways for discovering new gene biomarker combinations and previously unknown gene interactions.
- 2 March 2023
- Published by Wiley in American Journal of Medical Genetics Part B: Neuropsychiatric Genetics
- Vol. 192 (3-4) , 62-70
De novo mutations disturb early brain development more frequently than common variants in schizophrenia
Investigating functional, temporal, and cell‐type expression features of mutations is important for understanding a complex disease. Here, we collected and analyzed common variants and de novo mutations (DNMs) in schizophrenia (SCZ). We collected 2,636 missense and loss‐of‐function (LoF) DNMs in 2,263 genes across 3,477 SCZ patients (SCZ‐DNMs). We curated three gene lists: (a) SCZ‐neuroGenes (159 genes), which are intolerant to LoF and missense DNMs and are neurologically important, (b) SCZ‐moduleGenes (52 genes), which were derived from network analyses of SCZ‐DNMs, and (c) SCZ‐commonGenes (120 genes) from a recent GWAS as reference. To compare temporal gene expression, we used the BrainSpan dataset. We defined a fetal effect score (FES) to quantify the involvement of each gene in prenatal brain development. We further employed the specificity indexes (SIs) to evaluate cell‐type expression specificity from single‐cell expression data in cerebral cortices of humans and mice. Compared with SCZ‐commonGenes, SCZ‐neuroGenes and SCZ‐moduleGenes were highly expressed in the prenatal stage, had higher FESs, and had higher SIs in fetal replicating cells and undifferentiated cell types. Our results suggested that gene expression patterns in specific cell types in early fetal stages might have impacts on the risk of SCZ during adulthood.
- 2 March 2023
- Published by Wiley in American Journal of Medical Genetics Part B: Neuropsychiatric Genetics
- Vol. 192 (3-4) , 62-70
De novo mutations disturb early brain development more frequently than common variants in schizophrenia
Investigating functional, temporal, and cell‐type expression features of mutations is important for understanding a complex disease. Here, we collected and analyzed common variants and de novo mutations (DNMs) in schizophrenia (SCZ). We collected 2,636 missense and loss‐of‐function (LoF) DNMs in 2,263 genes across 3,477 SCZ patients (SCZ‐DNMs). We curated three gene lists: (a) SCZ‐neuroGenes (159 genes), which are intolerant to LoF and missense DNMs and are neurologically important, (b) SCZ‐moduleGenes (52 genes), which were derived from network analyses of SCZ‐DNMs, and (c) SCZ‐commonGenes (120 genes) from a recent GWAS as reference. To compare temporal gene expression, we used the BrainSpan dataset. We defined a fetal effect score (FES) to quantify the involvement of each gene in prenatal brain development. We further employed the specificity indexes (SIs) to evaluate cell‐type expression specificity from single‐cell expression data in cerebral cortices of humans and mice. Compared with SCZ‐commonGenes, SCZ‐neuroGenes and SCZ‐moduleGenes were highly expressed in the prenatal stage, had higher FESs, and had higher SIs in fetal replicating cells and undifferentiated cell types. Our results suggested that gene expression patterns in specific cell types in early fetal stages might have impacts on the risk of SCZ during adulthood.
- Open Access
- 10 October 2021
- Published by Springer Nature in Breast Cancer Research
- Vol. 23 (1) , 1-11
Artificial image objects for classification of breast cancer biomarkers with transcriptome sequencing data and convolutional neural network algorithms
Background: Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively for clinical applications due to the high dimension of the data and the highly correlated expression between individual genes.Methods: We proposed a method to transform RNA sequencing data into artificial image objects (AIOs) and applied convolutional neural network (CNN) algorithms to classify these AIOs. With the AIO technique, we considered each gene as a pixel in an image and its expression level as pixel intensity. Using the GSE96058 (n = 2976), GSE81538 (n = 405), and GSE163882 (n = 222) datasets, we created AIOs for the subjects and designed CNN models to classify biomarker Ki67 and Nottingham histologic grade (NHG).Results: With fivefold cross-validation, we accomplished a classification accuracy and AUC of 0.821 ± 0.023 and 0.891 ± 0.021 for Ki67 status. For NHG, the weighted average of categorical accuracy was 0.820 ± 0.012, and the weighted average of AUC was 0.931 ± 0.006. With GSE96058 as training data and GSE81538 as testing data, the accuracy and AUC for Ki67 were 0.826 ± 0.037 and 0.883 ± 0.016, and that for NHG were 0.764 ± 0.052 and 0.882 ± 0.012, respectively. These results were 10% better than the results reported in the original studies. For Ki67, the calls generated from our models had a better power for prediction of survival as compared to the calls from trained pathologists in survival analyses.Conclusions: We demonstrated that RNA sequencing data could be transformed into AIOs and be used to classify Ki67 status and NHG with CNN algorithms. The AIO method could handle high-dimensional data with highly correlated variables, and there was no need for variable selection. With the AIO technique, a data-driven, consistent, and automation-ready model could be developed to classify biomarkers with RNA sequencing data and provide more efficient care for cancer patients.
- 10 October 2021
- Published by Springer Nature in Breast Cancer Research
- Vol. 23 (1) , 1-11
Open AccessArtificial image objects for classification of breast cancer biomarkers with transcriptome sequencing data and convolutional neural network algorithms
Background: Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively for clinical applications due to the high dimension of the data and the highly correlated expression between individual genes.Methods: We proposed a method to transform RNA sequencing data into artificial image objects (AIOs) and applied convolutional neural network (CNN) algorithms to classify these AIOs. With the AIO technique, we considered each gene as a pixel in an image and its expression level as pixel intensity. Using the GSE96058 (n = 2976), GSE81538 (n = 405), and GSE163882 (n = 222) datasets, we created AIOs for the subjects and designed CNN models to classify biomarker Ki67 and Nottingham histologic grade (NHG).Results: With fivefold cross-validation, we accomplished a classification accuracy and AUC of 0.821 ± 0.023 and 0.891 ± 0.021 for Ki67 status. For NHG, the weighted average of categorical accuracy was 0.820 ± 0.012, and the weighted average of AUC was 0.931 ± 0.006. With GSE96058 as training data and GSE81538 as testing data, the accuracy and AUC for Ki67 were 0.826 ± 0.037 and 0.883 ± 0.016, and that for NHG were 0.764 ± 0.052 and 0.882 ± 0.012, respectively. These results were 10% better than the results reported in the original studies. For Ki67, the calls generated from our models had a better power for prediction of survival as compared to the calls from trained pathologists in survival analyses.Conclusions: We demonstrated that RNA sequencing data could be transformed into AIOs and be used to classify Ki67 status and NHG with CNN algorithms. The AIO method could handle high-dimensional data with highly correlated variables, and there was no need for variable selection. With the AIO technique, a data-driven, consistent, and automation-ready model could be developed to classify biomarkers with RNA sequencing data and provide more efficient care for cancer patients.
- Open Access
Rewired Pathways and Disrupted Pathway Crosstalk in Schizophrenia Transcriptomes by Multiple Differential Coexpression Methods
Transcriptomic studies of mental disorders using the human brain tissues have been limited, and gene expression signatures in schizophrenia (SCZ) remain elusive. In this study, we applied three differential co-expression methods to analyze five transcriptomic datasets (three RNA-Seq and two microarray datasets) derived from SCZ and matched normal postmortem brain samples. We aimed to uncover biological pathways where internal correlation structure was rewired or inter-coordination was disrupted in SCZ. In total, we identified 60 rewired pathways, many of which were related to neurotransmitter, synapse, immune, and cell adhesion. We found the hub genes, which were on the center of rewired pathways, were highly mutually consistent among the five datasets. The combinatory list of 92 hub genes was generally multi-functional, suggesting their complex and dynamic roles in SCZ pathophysiology. In our constructed pathway crosstalk network, we found “Clostridium neurotoxicity” and “signaling events mediated by focal adhesion kinase” had the highest interactions. We further identified disconnected gene links underlying the disrupted pathway crosstalk. Among them, four gene pairs (PAK1:SYT1, PAK1:RFC5, DCTN1:STX1A, and GRIA1:MAP2K4) were normally correlated in universal contexts. In summary, we systematically identified rewired pathways, disrupted pathway crosstalk circuits, and critical genes and gene links in schizophrenia transcriptomes.
Open AccessRewired Pathways and Disrupted Pathway Crosstalk in Schizophrenia Transcriptomes by Multiple Differential Coexpression Methods
Transcriptomic studies of mental disorders using the human brain tissues have been limited, and gene expression signatures in schizophrenia (SCZ) remain elusive. In this study, we applied three differential co-expression methods to analyze five transcriptomic datasets (three RNA-Seq and two microarray datasets) derived from SCZ and matched normal postmortem brain samples. We aimed to uncover biological pathways where internal correlation structure was rewired or inter-coordination was disrupted in SCZ. In total, we identified 60 rewired pathways, many of which were related to neurotransmitter, synapse, immune, and cell adhesion. We found the hub genes, which were on the center of rewired pathways, were highly mutually consistent among the five datasets. The combinatory list of 92 hub genes was generally multi-functional, suggesting their complex and dynamic roles in SCZ pathophysiology. In our constructed pathway crosstalk network, we found “Clostridium neurotoxicity” and “signaling events mediated by focal adhesion kinase” had the highest interactions. We further identified disconnected gene links underlying the disrupted pathway crosstalk. Among them, four gene pairs (PAK1:SYT1, PAK1:RFC5, DCTN1:STX1A, and GRIA1:MAP2K4) were normally correlated in universal contexts. In summary, we systematically identified rewired pathways, disrupted pathway crosstalk circuits, and critical genes and gene links in schizophrenia transcriptomes.