Xiaoman Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaoman Li has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Haiyan Hu
University of Central Florida
56 shared publications Saidi Wang
Department of Computer Science, University of Central Florida, Orlando, Florida, United States of America
18 shared publicationsAmlan Talukder
Department of Computer Science, University of Central Florida, Orlando, USA
11 shared publicationsHaiyan Hu
Department of Computer Science, University of Central Florida, Orlando, Florida, USA
10 shared publicationsHansi Zheng
Department of Computer Science, University of Central Florida, Orlando, Florida, United States of America
8 shared publicationsRony Chowdhury Ripan
Department of Computer Science, University of Central Florida, Orlando, USA
6 shared publications
Latest Publications
- Open Access
- 2 November 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (11) , e0360620
Recurrent enhancer-promoter interactions across samples
Enhancer-promoter interactions (EPIs) are fundamental to gene regulation, and clarifying their recurrence across diverse biological contexts is key to understanding chromatin architecture. Here, we systematically evaluate EPI recurrence using 95 HiChIP and 266 Hi-C datasets spanning multiple cell types, cell lines, and experimental conditions. We find that each of the majority of EPIs identified in any given sample is also observed in other samples, largely independent of assay type, cellular context, or enhancer annotation strategy. EPIs unique to an individual sample tend to occur in regions with fewer neighboring EPIs or in a sample with greater sequencing depth, indicating that many such “unique” EPIs likely reflect low sequencing depth rather than true biological specificity. Collectively, our results suggest that most human EPIs may have already been cataloged, and that gene regulation primarily relies on the selective reuse of a shared pool of EPIs across contexts. Thus, while EPIs are context-specific in pairwise sample comparisons, they are broadly recurrent when considered across a large compendium of samples. These findings underscore the pervasive and reusable nature of EPIs and have important implications for the interpretation of chromatin conformation data.
- 2 November 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (11) , e0360620
Open AccessRecurrent enhancer-promoter interactions across samples
Enhancer-promoter interactions (EPIs) are fundamental to gene regulation, and clarifying their recurrence across diverse biological contexts is key to understanding chromatin architecture. Here, we systematically evaluate EPI recurrence using 95 HiChIP and 266 Hi-C datasets spanning multiple cell types, cell lines, and experimental conditions. We find that each of the majority of EPIs identified in any given sample is also observed in other samples, largely independent of assay type, cellular context, or enhancer annotation strategy. EPIs unique to an individual sample tend to occur in regions with fewer neighboring EPIs or in a sample with greater sequencing depth, indicating that many such “unique” EPIs likely reflect low sequencing depth rather than true biological specificity. Collectively, our results suggest that most human EPIs may have already been cataloged, and that gene regulation primarily relies on the selective reuse of a shared pool of EPIs across contexts. Thus, while EPIs are context-specific in pairwise sample comparisons, they are broadly recurrent when considered across a large compendium of samples. These findings underscore the pervasive and reusable nature of EPIs and have important implications for the interpretation of chromatin conformation data.
- Open Access
- 16 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
misoTar: a novel approach for predicting miRNA and isomiR targets
Understanding microRNA/isomiR–mRNA interactions has long been a major challenge. Although many computational methods exist for predicting miRNA–mRNA interactions, almost all fail to account for isomiR–mRNA interactions. To bridge this gap, we developed misoTar, a fine-tuned BERT-based deep learning model trained on over 6.660 million positive and negative microRNA/isomiR–mRNA interaction pairs from 67 publicly available human samples across six studies. In five-fold cross-validation, misoTar achieved an average precision of 0.930 and a recall of 0.898. On independent test datasets, it consistently delivered superior or comparable performance relative to existing tools, including TargetScan, Mimosa, DMISO, and TEC-miTarget. Furthermore, single-nucleotide mutation analysis of true positive interactions highlighted the critical functional importance of non-seed regions in microRNA/isomiR–mRNA targeting. Overall, misoTar offers a robust and accurate framework for predicting microRNA/isomiR–mRNA interactions and provides new insights into microRNA biology. The misoTar tool is available at https://figshare.com/projects/misoTar/262723.
- 16 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessmisoTar: a novel approach for predicting miRNA and isomiR targets
Understanding microRNA/isomiR–mRNA interactions has long been a major challenge. Although many computational methods exist for predicting miRNA–mRNA interactions, almost all fail to account for isomiR–mRNA interactions. To bridge this gap, we developed misoTar, a fine-tuned BERT-based deep learning model trained on over 6.660 million positive and negative microRNA/isomiR–mRNA interaction pairs from 67 publicly available human samples across six studies. In five-fold cross-validation, misoTar achieved an average precision of 0.930 and a recall of 0.898. On independent test datasets, it consistently delivered superior or comparable performance relative to existing tools, including TargetScan, Mimosa, DMISO, and TEC-miTarget. Furthermore, single-nucleotide mutation analysis of true positive interactions highlighted the critical functional importance of non-seed regions in microRNA/isomiR–mRNA targeting. Overall, misoTar offers a robust and accurate framework for predicting microRNA/isomiR–mRNA interactions and provides new insights into microRNA biology. The misoTar tool is available at https://figshare.com/projects/misoTar/262723.
- Open Access
Preferential IsomiR Enrichment in Extracellular Vesicles Improves Identification of Their Cellular Origins
Extracellular vesicles (EVs) carry microRNAs (miRNAs) that mediate intercellular communication and have strong potential as disease biomarkers, yet the roles of miRNA isoforms (isomiRs) in EVs remain poorly understood. Here, we analyzed 96 human EV and corresponding source samples from nine public datasets. We found that EV samples consistently contained substantially higher proportions of isomiR reads than their corresponding source samples, indicating widespread isomiR enrichment in EVs. Although individual isomiRs showed limited reproducibility across biological replicates and limited sharing between EVs and their corresponding source samples, the parent miRNAs that generated these isomiRs remained highly reproducible across replicates and strongly shared between EV-source pairs. Despite extensive isomiR diversification, EV-source pairs retained highly correlated miRNA expression profiles. Using integrated miRNA- and isomiR-related features, we further developed a random forest model that successfully associated EV samples with their corresponding source samples, with improved performance when isomiR information was included. Together, our results demonstrate that EVs are enriched for biologically meaningful isomiRs while preserving source-associated miRNA landscapes, highlighting the importance of incorporating isomiRs into future EV studies.
Open AccessPreferential IsomiR Enrichment in Extracellular Vesicles Improves Identification of Their Cellular Origins
Extracellular vesicles (EVs) carry microRNAs (miRNAs) that mediate intercellular communication and have strong potential as disease biomarkers, yet the roles of miRNA isoforms (isomiRs) in EVs remain poorly understood. Here, we analyzed 96 human EV and corresponding source samples from nine public datasets. We found that EV samples consistently contained substantially higher proportions of isomiR reads than their corresponding source samples, indicating widespread isomiR enrichment in EVs. Although individual isomiRs showed limited reproducibility across biological replicates and limited sharing between EVs and their corresponding source samples, the parent miRNAs that generated these isomiRs remained highly reproducible across replicates and strongly shared between EV-source pairs. Despite extensive isomiR diversification, EV-source pairs retained highly correlated miRNA expression profiles. Using integrated miRNA- and isomiR-related features, we further developed a random forest model that successfully associated EV samples with their corresponding source samples, with improved performance when isomiR information was included. Together, our results demonstrate that EVs are enriched for biologically meaningful isomiRs while preserving source-associated miRNA landscapes, highlighting the importance of incorporating isomiRs into future EV studies.
- Open Access
misoTar: A novel approach for predicting miRNA and isomiR targets
Understanding the interactions between microRNAs/isomiRs and mRNAs has long been a major challenge in RNA biology. Although numerous computational approaches have been developed to predict these interactions, most fail to account for isomiR mediated targeting. To address this limitation, we developed misoTar, a deep learning framework trained on more than 6.662 million positive and negative interaction pairs derived from 67 publicly available human samples across six independent studies. In five-fold cross-validation, misoTar achieved an average precision of 0.930 and a recall of 0.898. Evaluation on independent test datasets demonstrated consistently superior or comparable performance relative to existing tools, including TargetScan, Mimosa, DMISO, and TEC-miTarget. In addition, single-nucleotide mutation analyses of true positive interactions revealed the critical functional contributions of non-seed regions in microRNA/isomiR targeting. Overall, misoTar provides a robust and accurate framework for predicting microRNA/isomiR interactions while offering new biological insights into microRNA targeting mechanisms. The misoTar tool is publicly available at https://figshare.com/projects/misoTar/262723 .
Open AccessmisoTar: A novel approach for predicting miRNA and isomiR targets
Understanding the interactions between microRNAs/isomiRs and mRNAs has long been a major challenge in RNA biology. Although numerous computational approaches have been developed to predict these interactions, most fail to account for isomiR mediated targeting. To address this limitation, we developed misoTar, a deep learning framework trained on more than 6.662 million positive and negative interaction pairs derived from 67 publicly available human samples across six independent studies. In five-fold cross-validation, misoTar achieved an average precision of 0.930 and a recall of 0.898. Evaluation on independent test datasets demonstrated consistently superior or comparable performance relative to existing tools, including TargetScan, Mimosa, DMISO, and TEC-miTarget. In addition, single-nucleotide mutation analyses of true positive interactions revealed the critical functional contributions of non-seed regions in microRNA/isomiR targeting. Overall, misoTar provides a robust and accurate framework for predicting microRNA/isomiR interactions while offering new biological insights into microRNA targeting mechanisms. The misoTar tool is publicly available at https://figshare.com/projects/misoTar/262723 .
- Open Access
- 24 April 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
IST: an ontology-guided attention-based autoencoder for interpretable analysis of single-cell transcriptomic data
Extracting biologically interpretable insights from single-cell RNA sequencing (scRNA-seq) data remains a major challenge. While deep learning approaches have demonstrated strong predictive performance for dimensionality reduction and representation learning, their latent representations often lack clear biological meaning, and prior biological knowledge is frequently incorporated only weakly or inconsistently. As a result, learned features may not faithfully reflect known gene-function relationships, limiting biological interpretability and downstream insight. We present IST (Interpretation of Single-cell Transcriptomic data), a novel attention-based framework designed to reliably integrate biological prior knowledge into the analysis of single-cell transcriptomic data. IST incorporates gene ontology structure directly into the model architecture and employs attention mechanisms to capture context-dependent gene activity. A tailored correlation-based loss enforces alignment between learned gene activities and known gene-function relationships, ensuring that biological priors are faithfully reflected in the learned representations rather than acting as superficial regularization. We evaluate IST on three scRNA-seq datasets. Across all datasets, IST identifies condition-specific gene activity patterns, uncovers novel relationships among biological processes, and infers previously unannotated gene functions supported by existing literature. Compared with state-of-the-art methods, IST provides more interpretable and biologically grounded representations of gene functions and activities, while maintaining competitive predictive performance. The datasets and tool are available at 10.6084/m9.figshare.30811334.
- 24 April 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessIST: an ontology-guided attention-based autoencoder for interpretable analysis of single-cell transcriptomic data
Extracting biologically interpretable insights from single-cell RNA sequencing (scRNA-seq) data remains a major challenge. While deep learning approaches have demonstrated strong predictive performance for dimensionality reduction and representation learning, their latent representations often lack clear biological meaning, and prior biological knowledge is frequently incorporated only weakly or inconsistently. As a result, learned features may not faithfully reflect known gene-function relationships, limiting biological interpretability and downstream insight. We present IST (Interpretation of Single-cell Transcriptomic data), a novel attention-based framework designed to reliably integrate biological prior knowledge into the analysis of single-cell transcriptomic data. IST incorporates gene ontology structure directly into the model architecture and employs attention mechanisms to capture context-dependent gene activity. A tailored correlation-based loss enforces alignment between learned gene activities and known gene-function relationships, ensuring that biological priors are faithfully reflected in the learned representations rather than acting as superficial regularization. We evaluate IST on three scRNA-seq datasets. Across all datasets, IST identifies condition-specific gene activity patterns, uncovers novel relationships among biological processes, and infers previously unannotated gene functions supported by existing literature. Compared with state-of-the-art methods, IST provides more interpretable and biologically grounded representations of gene functions and activities, while maintaining competitive predictive performance. The datasets and tool are available at 10.6084/m9.figshare.30811334.
- Open Access
- 28 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-9
New characteristics of MiRNA and IsomiR interactions with mRNA
Studying the interactions between microRNAs/isomiRs and mRNAs is crucial due to their fundamental roles in gene regulation and disease. Although many isomiRs have been identified, the analysis of their interactions with mRNAs remains in its early stages. In this study, we compiled available human chimeric reads, each pairing a microRNA or isomiR segment with an mRNA segment. We then identified 1747 isomiRs and over 5 million microRNA/isomiR–mRNA interactions from the complied data. We found that microRNAs with higher adenine and thymine content, and lower cytosine content, tend to have more isomiRs and target more mRNAs. Notably, 18.9% of mRNA targets were bound exclusively by isomiRs, not their microRNAs. Furthermore, isomiRs sharing the same seed sequences as their reference miRNAs may target different mRNAs from their miRNAs, suggesting functional divergence. Interestingly, 20.0% of microRNAs and 8.2% of isomiRs bind mRNAs independently of their seed regions. Among those that do utilize seed regions, 94.5% of microRNAs and 95.7% of isomiRs also engage non-seed region binding. Our findings provide new insights into the complexity of microRNA/isomiR–mRNA interactions.
- 28 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-9
Open AccessNew characteristics of MiRNA and IsomiR interactions with mRNA
Studying the interactions between microRNAs/isomiRs and mRNAs is crucial due to their fundamental roles in gene regulation and disease. Although many isomiRs have been identified, the analysis of their interactions with mRNAs remains in its early stages. In this study, we compiled available human chimeric reads, each pairing a microRNA or isomiR segment with an mRNA segment. We then identified 1747 isomiRs and over 5 million microRNA/isomiR–mRNA interactions from the complied data. We found that microRNAs with higher adenine and thymine content, and lower cytosine content, tend to have more isomiRs and target more mRNAs. Notably, 18.9% of mRNA targets were bound exclusively by isomiRs, not their microRNAs. Furthermore, isomiRs sharing the same seed sequences as their reference miRNAs may target different mRNAs from their miRNAs, suggesting functional divergence. Interestingly, 20.0% of microRNAs and 8.2% of isomiRs bind mRNAs independently of their seed regions. Among those that do utilize seed regions, 94.5% of microRNAs and 95.7% of isomiRs also engage non-seed region binding. Our findings provide new insights into the complexity of microRNA/isomiR–mRNA interactions.
- Open Access
Recurrent enhancer-promoter interactions across samples
Enhancer-promoter interactions (EPIs) are fundamental to gene regulation, and understanding their recurrence across diverse biological samples is key to deciphering chromatin architecture. In this study, we systematically analyzed the recurrence of EPIs across 49 Hi-C and 95 HiChIP datasets. We found that the majority of EPIs identified in a given sample were also present in other samples, regardless of the assay type (Hi-C or HiChIP) or the enhancer annotations used. Interestingly, EPIs that appeared unique to individual samples were typically surrounded by fewer neighboring EPIs, suggesting they may not represent truly sample-specific interactions. Our findings indicate that most human EPIs have already been captured and that cells primarily reuse subsets of these shared EPIs across different cell types and conditions. This study provides new insights into the pervasive and reusable nature of EPIs in the human genome, with important implications for chromatin conformation studies.
Open AccessRecurrent enhancer-promoter interactions across samples
Enhancer-promoter interactions (EPIs) are fundamental to gene regulation, and understanding their recurrence across diverse biological samples is key to deciphering chromatin architecture. In this study, we systematically analyzed the recurrence of EPIs across 49 Hi-C and 95 HiChIP datasets. We found that the majority of EPIs identified in a given sample were also present in other samples, regardless of the assay type (Hi-C or HiChIP) or the enhancer annotations used. Interestingly, EPIs that appeared unique to individual samples were typically surrounded by fewer neighboring EPIs, suggesting they may not represent truly sample-specific interactions. Our findings indicate that most human EPIs have already been captured and that cells primarily reuse subsets of these shared EPIs across different cell types and conditions. This study provides new insights into the pervasive and reusable nature of EPIs in the human genome, with important implications for chromatin conformation studies.
- 4 July 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Bioinformatics and Computational Biology
- Vol. 23 (3) , 2550009
Deep learning inference of mirna expression from bulk and single-cell mrna expression
Studying miRNA activity at the single-cell level presents a significant challenge due to the limitations of existing single-cell technologies in capturing miRNAs. To address this, we introduce two deep learning models: Cross-modality (CM) and single-modality (SM), both based on encoder-decoder architectures. These models predict miRNA expression at both bulk and single-cell levels using mRNA data. We evaluated the performance of CM and SM against the state-of-the-art miRSCAPE approach, using both bulk and single-cell datasets. Our results demonstrate that both CM and SM outperform miRSCAPE in accuracy. Furthermore, incorporating miRNA target information substantially enhanced performance compared to models that utilized all genes. These models provide powerful tools for predicting miRNA expression from single-cell mRNA data.
- 4 July 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Bioinformatics and Computational Biology
- Vol. 23 (3) , 2550009
Deep learning inference of mirna expression from bulk and single-cell mrna expression
Studying miRNA activity at the single-cell level presents a significant challenge due to the limitations of existing single-cell technologies in capturing miRNAs. To address this, we introduce two deep learning models: Cross-modality (CM) and single-modality (SM), both based on encoder-decoder architectures. These models predict miRNA expression at both bulk and single-cell levels using mRNA data. We evaluated the performance of CM and SM against the state-of-the-art miRSCAPE approach, using both bulk and single-cell datasets. Our results demonstrate that both CM and SM outperform miRSCAPE in accuracy. Furthermore, incorporating miRNA target information substantially enhanced performance compared to models that utilized all genes. These models provide powerful tools for predicting miRNA expression from single-cell mRNA data.
- Open Access
Novel features of miRNA and isomiR-mRNA interactions
Studying the interactions between microRNAs/isomiRs and mRNAs is critical due to their fundamental roles in gene regulation and their involvement in various diseases. Although many isomiRs have been identified, the analysis of their interactions with mRNAs remains in its early stages. In this study, we compiled available human chimeric read data, each comprising a microRNA or isomiR segment paired with an mRNA fragment and identified 1,747 isomiRs and over 5 million microRNA/isomiR–mRNA interactions. We observed that microRNAs with higher adenine and thymine content, and lower cytosine content, tend to have more isomiRs and more mRNA targets. Notably, an average of 18.9% of mRNA targets were bound exclusively by isomiRs, not by their microRNA counterparts. Furthermore, isomiRs sharing the same seed sequences as their reference microRNAs may bind to different targets from their microRNAs, highlighting functional divergence. Interestingly, 20.0% of microRNAs and 8.2% of isomiRs appear to bind mRNAs independently of their seed regions. Among those that do utilize seed regions, 94.5% of microRNAs and 95.7% of isomiRs also engage non-seed regions, suggesting a broader and more complex binding behavior. Our findings offer new insights into microRNA/isomiR–mRNA interactions.
Open AccessNovel features of miRNA and isomiR-mRNA interactions
Studying the interactions between microRNAs/isomiRs and mRNAs is critical due to their fundamental roles in gene regulation and their involvement in various diseases. Although many isomiRs have been identified, the analysis of their interactions with mRNAs remains in its early stages. In this study, we compiled available human chimeric read data, each comprising a microRNA or isomiR segment paired with an mRNA fragment and identified 1,747 isomiRs and over 5 million microRNA/isomiR–mRNA interactions. We observed that microRNAs with higher adenine and thymine content, and lower cytosine content, tend to have more isomiRs and more mRNA targets. Notably, an average of 18.9% of mRNA targets were bound exclusively by isomiRs, not by their microRNA counterparts. Furthermore, isomiRs sharing the same seed sequences as their reference microRNAs may bind to different targets from their microRNAs, highlighting functional divergence. Interestingly, 20.0% of microRNAs and 8.2% of isomiRs appear to bind mRNAs independently of their seed regions. Among those that do utilize seed regions, 94.5% of microRNAs and 95.7% of isomiRs also engage non-seed regions, suggesting a broader and more complex binding behavior. Our findings offer new insights into microRNA/isomiR–mRNA interactions.
- Open Access
Deep learning inference of miRNA expression from bulk and single-cell mRNA expression
Understanding the activity of miRNA in individual cells presents a challenge due to the limitations of single-cell technologies in capturing miRNAs. To tackle this obstacle, we introduce two deep learning models: Cross-Modality (CM) and Single-Modality (SM). These models utilize encoder-decoder architectures to predict miRNA expression at the bulk and single-cell levels from mRNA data. We compared CM and SM with a state-of-the-art approach, miRSCAPE, using both bulk and single-cell datasets. We found that both CM and SM outperformed miRSCAPE in terms of accuracy. We also observed that integrating miRNA target information led to a significant enhancement in performance compared to using all genes. These models offer valuable tools for predicting miRNA expression from single-cell mRNA data.
Open AccessDeep learning inference of miRNA expression from bulk and single-cell mRNA expression
Understanding the activity of miRNA in individual cells presents a challenge due to the limitations of single-cell technologies in capturing miRNAs. To tackle this obstacle, we introduce two deep learning models: Cross-Modality (CM) and Single-Modality (SM). These models utilize encoder-decoder architectures to predict miRNA expression at the bulk and single-cell levels from mRNA data. We compared CM and SM with a state-of-the-art approach, miRSCAPE, using both bulk and single-cell datasets. We found that both CM and SM outperformed miRSCAPE in terms of accuracy. We also observed that integrating miRNA target information led to a significant enhancement in performance compared to using all genes. These models offer valuable tools for predicting miRNA expression from single-cell mRNA data.