Li Shen
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
Li Shen has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Bairong Shen
Department of Respiratory and Critical Care Medicine and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu, China
3 shared publications - Rongrong Wu
Joint Laboratory of Artificial Intelligence for Critical Care Medicine, Department of Critical Care Medicine and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu, China
3 shared publications - Shumin Ren
Institutes for Systems Genetics, West China Hospital of Sichuan University, Frontiers Science Center for Disease-related Molecular Network, Chengdu, Sichuan, 610041, China, 86 15995854635
3 shared publications - Xingyun Liu
Affiliations Health Management Center, General Practice Medical Center and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu, China, Department of Computer Science and Information Technology, University of A Coruña, Coruña, Spain
3 shared publications - Tong Tang
Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu 610041, China; Department of Computer Science and Information Technologies, Elviña Campus, University of A Coruña, 15001 A Coruña, Spain
2 shared publications Ali Hashmi
University of Helsinki
1 shared publication
Latest Publications
- 30 May 2026
- Published by Wiley in Advanced Biology
- Vol. 10 (6) , e70129
Integrative Database‐Driven In Silico and In Vitro Study of Anemarrhena asphodeloides Bunge Highlighting Hippeastrine as a Regulator of the HSP90/PI3K/Akt/mTOR Axis in Oral Squamous Cell Carcinoma
Oral squamous cell carcinoma (OSCC) has a poor prognosis and limited effective therapies. Anemarrhena asphodeloides exhibits anticancer activity, but its efficacy in OSCC treatment remains underexplored. Network and protein‐protein interaction analysis identified 275 overlapping targets and 12 hub genes, with HSP90AA1, HSP90AB1, and CASP3 upregulated in OSCC, linked to poorer survival. QSAR modeling, molecular docking, and molecular dynamics simulations demonstrated stable binding of key bioactive compounds with HSP90 and AKT1, highlighting hippeastrine as a promising candidate. ADMET analysis indicated their favorable pharmacokinetic profiles. To validate the therapeutic potential of hippeastrine, in vitro experiments were conducted using the CAL27 and SCC‐9 cell lines. CCK‐8 assay demonstrated that hippeastrine exerted significant and dose‐dependent cytotoxicity against OSCC cells. Furthermore, hippeastrine treatment significantly suppressed colony formation and attenuated cell migration and invasion capabilities. Gene set enrichment analysis of TCGA‐HNSC data revealed a positive correlation between HSP90AA1/AB1 expression and PI3K/Akt/mTOR signaling. Western blot analysis confirmed inhibition of the HSP90‐PI3K‐Akt‐mTOR axis by hippeastrine. Moreover, functional rescue experiments with the PI3K activator 740 Y‐P successfully reversed the hippeastrine‐induced suppression of cell viability and PI3K/Akt phosphorylation. These findings indicate that hippeastrine exerts anti‐OSCC effects by targeting the HSP90/PI3K/Akt/mTOR pathway, highlighting it as a novel therapeutic agent against OSCC.
- 30 May 2026
- Published by Wiley in Advanced Biology
- Vol. 10 (6) , e70129
Integrative Database‐Driven In Silico and In Vitro Study of Anemarrhena asphodeloides Bunge Highlighting Hippeastrine as a Regulator of the HSP90/PI3K/Akt/mTOR Axis in Oral Squamous Cell Carcinoma
Oral squamous cell carcinoma (OSCC) has a poor prognosis and limited effective therapies. Anemarrhena asphodeloides exhibits anticancer activity, but its efficacy in OSCC treatment remains underexplored. Network and protein‐protein interaction analysis identified 275 overlapping targets and 12 hub genes, with HSP90AA1, HSP90AB1, and CASP3 upregulated in OSCC, linked to poorer survival. QSAR modeling, molecular docking, and molecular dynamics simulations demonstrated stable binding of key bioactive compounds with HSP90 and AKT1, highlighting hippeastrine as a promising candidate. ADMET analysis indicated their favorable pharmacokinetic profiles. To validate the therapeutic potential of hippeastrine, in vitro experiments were conducted using the CAL27 and SCC‐9 cell lines. CCK‐8 assay demonstrated that hippeastrine exerted significant and dose‐dependent cytotoxicity against OSCC cells. Furthermore, hippeastrine treatment significantly suppressed colony formation and attenuated cell migration and invasion capabilities. Gene set enrichment analysis of TCGA‐HNSC data revealed a positive correlation between HSP90AA1/AB1 expression and PI3K/Akt/mTOR signaling. Western blot analysis confirmed inhibition of the HSP90‐PI3K‐Akt‐mTOR axis by hippeastrine. Moreover, functional rescue experiments with the PI3K activator 740 Y‐P successfully reversed the hippeastrine‐induced suppression of cell viability and PI3K/Akt phosphorylation. These findings indicate that hippeastrine exerts anti‐OSCC effects by targeting the HSP90/PI3K/Akt/mTOR pathway, highlighting it as a novel therapeutic agent against OSCC.
- Open Access
Predicting Pre-treatment Resistance or Post-treatment Effect? A Systematic Benchmarking of Single-Cell Drug Response Models
Intratumoral heterogeneity drives variable drug responses in cancer. Single-cell RNA sequencing (scRNA-seq) enables characterization of such heterogeneity and prediction of drug response at single-cell resolution. Accordingly, various computational models have been developed to infer drug response from scRNA-seq data. However, their performance, robustness, and generalizability across different biological contexts remain insufficiently evaluated. To address this gap, we benchmarked representative single-cell drug response prediction models using 26 curated datasets comprising over 760,000 cells across 12 cancer types and 21 therapeutic agents. We constructed balanced and imbalanced scenarios to explore model behavior under different label distributions. To address the lack of ground-truth labels in conventional scRNA-seq datasets, we incorporated two lineage-tracing datasets with experimentally validated drug-response annotations, enabling evaluation in a clinically relevant pre-treatment prediction setting. Our results show that prediction performance was markedly higher in cell lines than in tissue samples and varied substantially across datasets, treatment types, and imbalance levels. Under increasing class imbalance, several methods showed reduced performance, and no single method consistently dominated across all evaluation settings. In lineage-tracing datasets, treatment-associated states were more readily distinguished than lineage-defined pre-sensitive and pre-resistant states, while transcriptomic analyses indicated that pre-existing resistance-related signals were detectable but relatively subtle. For scDEAL, which showed competitive performance across several benchmark settings, label-substitution and ablation analyses further indicated that its performance was jointly influenced by training-label definition and architectural design. Application to an independent PDAC organoid dataset showed that scDEAL predictions were associated with treatment-related transcriptional state changes and biological programs. In summary, our study systematically characterizes the performance and limitations of current single-cell drug response prediction approaches across diverse evaluation settings. These results highlight the influence of dataset characteristics, class imbalance, training design, and the strength of resistance-associated signals on model performance, providing a reference for the evaluation and further development of single-cell drug response prediction methods.
Open AccessPredicting Pre-treatment Resistance or Post-treatment Effect? A Systematic Benchmarking of Single-Cell Drug Response Models
Intratumoral heterogeneity drives variable drug responses in cancer. Single-cell RNA sequencing (scRNA-seq) enables characterization of such heterogeneity and prediction of drug response at single-cell resolution. Accordingly, various computational models have been developed to infer drug response from scRNA-seq data. However, their performance, robustness, and generalizability across different biological contexts remain insufficiently evaluated. To address this gap, we benchmarked representative single-cell drug response prediction models using 26 curated datasets comprising over 760,000 cells across 12 cancer types and 21 therapeutic agents. We constructed balanced and imbalanced scenarios to explore model behavior under different label distributions. To address the lack of ground-truth labels in conventional scRNA-seq datasets, we incorporated two lineage-tracing datasets with experimentally validated drug-response annotations, enabling evaluation in a clinically relevant pre-treatment prediction setting. Our results show that prediction performance was markedly higher in cell lines than in tissue samples and varied substantially across datasets, treatment types, and imbalance levels. Under increasing class imbalance, several methods showed reduced performance, and no single method consistently dominated across all evaluation settings. In lineage-tracing datasets, treatment-associated states were more readily distinguished than lineage-defined pre-sensitive and pre-resistant states, while transcriptomic analyses indicated that pre-existing resistance-related signals were detectable but relatively subtle. For scDEAL, which showed competitive performance across several benchmark settings, label-substitution and ablation analyses further indicated that its performance was jointly influenced by training-label definition and architectural design. Application to an independent PDAC organoid dataset showed that scDEAL predictions were associated with treatment-related transcriptional state changes and biological programs. In summary, our study systematically characterizes the performance and limitations of current single-cell drug response prediction approaches across diverse evaluation settings. These results highlight the influence of dataset characteristics, class imbalance, training design, and the strength of resistance-associated signals on model performance, providing a reference for the evaluation and further development of single-cell drug response prediction methods.
- Open Access
- 10 December 2024
- Published by JMIR Publications Inc. in JMIR Preprints
PEP-DSS-DBC1.0: A Personalized Exercise Prescription Decision Support System for Diabetes Care (Preprint)
BACKGROUND Exercise plays a critical role in managing diabetes, improving glycemic control, and reducing the risk of cardiovascular complications. However, personalizing exercise prescriptions (PEPs) to address individual characteristics and preferences is challenging, particularly in an environment with a global shortage of exercise professionals. While digital tools hold promise for improving personalized care, many existing systems rely on generalized guidelines or expert consensus, lacking the ability to provide highly individualized and evidence-based recommendations. To address this gap, we developed PEP-DSS-DBC1.0, a knowledge-based digital decision support system for generating tailored PEPs in diabetes care. OBJECTIVE The aim of this study was to create a digital, knowledge-based decision support tool for personalized exercise prescriptions that incorporates individual characteristics and preferences. PEP-DSS-DBC1.0 was designed to enhance the precision and usability of PEPs for people with diabetes while ensuring alignment with evidence-based principles. METHODS The development of PEP-DSS-DBC1.0 was informed by an extensive review of the literature. A working group, including experts in exercise medicine, diabetes care, and informatics, was formed to establish a decision support framework. The PubMed database served as the primary resource for knowledge extraction, covering studies published between January 1975 and May 2023. Data from 938 studies were annotated, resulting in 1,171 exercise protocols, 10,316,497 patient records, and 2,904 additional entries, totaling 13,263 data points. The web-based tool was built using a browser/server architecture and the WIMA (Windows, IIS, MySQL, ASP.NET) web stack. Validation and comparative analysis of the tool were performed against ChatGPT4o and sports physicians, using multiple metrics and two rating scales: the System Usability Scale (SUS) and the Net Promoter Score (NPS). RESULTS PEP-DSS-DBC1.0 provides personalized exercise prescriptions based on the newly developed “FITT-VP-WC” principles, which include frequency, intensity, time, type, volume, progression, warning, and contraindication. The tool demonstrated superior performance compared to ChatGPT4o and sports physicians, achieving a total performance score of 100% compared to 48% for ChatGPT4o and 12% for sports physicians (χ2=853.3, P<.001). Usability assessments showed that the average SUS score was 77.25 ± 6.06, indicating above-average usability, while the NPS score was 68.75%, reflecting high user loyalty and satisfaction. The tool is freely accessible at http://PEP-DSS-DBC.sysbio.org.cn/. CONCLUSIONS PEP-DSS-DBC1.0 represents a significant advancement in diabetes care, providing a knowledge-based system for generating highly individualized exercise prescriptions. While the tool has demonstrated superior performance and usability, future randomized controlled trials are essential to further validate its clinical utility and broader application in diabetes management.
- 10 December 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPEP-DSS-DBC1.0: A Personalized Exercise Prescription Decision Support System for Diabetes Care (Preprint)
BACKGROUND Exercise plays a critical role in managing diabetes, improving glycemic control, and reducing the risk of cardiovascular complications. However, personalizing exercise prescriptions (PEPs) to address individual characteristics and preferences is challenging, particularly in an environment with a global shortage of exercise professionals. While digital tools hold promise for improving personalized care, many existing systems rely on generalized guidelines or expert consensus, lacking the ability to provide highly individualized and evidence-based recommendations. To address this gap, we developed PEP-DSS-DBC1.0, a knowledge-based digital decision support system for generating tailored PEPs in diabetes care. OBJECTIVE The aim of this study was to create a digital, knowledge-based decision support tool for personalized exercise prescriptions that incorporates individual characteristics and preferences. PEP-DSS-DBC1.0 was designed to enhance the precision and usability of PEPs for people with diabetes while ensuring alignment with evidence-based principles. METHODS The development of PEP-DSS-DBC1.0 was informed by an extensive review of the literature. A working group, including experts in exercise medicine, diabetes care, and informatics, was formed to establish a decision support framework. The PubMed database served as the primary resource for knowledge extraction, covering studies published between January 1975 and May 2023. Data from 938 studies were annotated, resulting in 1,171 exercise protocols, 10,316,497 patient records, and 2,904 additional entries, totaling 13,263 data points. The web-based tool was built using a browser/server architecture and the WIMA (Windows, IIS, MySQL, ASP.NET) web stack. Validation and comparative analysis of the tool were performed against ChatGPT4o and sports physicians, using multiple metrics and two rating scales: the System Usability Scale (SUS) and the Net Promoter Score (NPS). RESULTS PEP-DSS-DBC1.0 provides personalized exercise prescriptions based on the newly developed “FITT-VP-WC” principles, which include frequency, intensity, time, type, volume, progression, warning, and contraindication. The tool demonstrated superior performance compared to ChatGPT4o and sports physicians, achieving a total performance score of 100% compared to 48% for ChatGPT4o and 12% for sports physicians (χ2=853.3, P<.001). Usability assessments showed that the average SUS score was 77.25 ± 6.06, indicating above-average usability, while the NPS score was 68.75%, reflecting high user loyalty and satisfaction. The tool is freely accessible at http://PEP-DSS-DBC.sysbio.org.cn/. CONCLUSIONS PEP-DSS-DBC1.0 represents a significant advancement in diabetes care, providing a knowledge-based system for generating highly individualized exercise prescriptions. While the tool has demonstrated superior performance and usability, future randomized controlled trials are essential to further validate its clinical utility and broader application in diabetes management.
- Open Access
- 17 October 2022
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 21 (2) , 292-299
CTRR-ncRNA: A Knowledgebase for Cancer Therapy Resistance and Recurrence Associated Non-Coding RNAs
Cancer therapy resistance and recurrence (CTRR) are the dominant causes of death in cancer patients. Recent studies have indicated that non-coding RNAs (ncRNAs) can not only reverse the resistance to cancer therapy but also are crucial biomarkers for the evaluation and prediction of CTRR. Herein, we developed CTRR-ncRNA, a knowledgebase of CTRR-associated ncRNAs, aiming to provide an accurate and comprehensive resource for research involving the association between CTRR and ncRNAs. Compared to most of the existing cancer databases, CTRR-ncRNA is focused on the clinical characterization of cancers, including cancer subtypes, as well as survival outcomes and responses to personalized therapy of cancer patients. Information pertaining to biomarker ncRNAs has also been documented for the development of personalized CTRR prediction. A user-friendly interface and several functional modules have been incorporated into the database. Based on the preliminary analysis of genotype–phenotype relationships, universal ncRNAs have been found to be potential biomarkers for CTRR. The CTRR-ncRNA is a translation-oriented knowledgebase and it provides a valuable resource for mechanistic investigations and explainable artificial intelligence-based modeling. CTRR-ncRNA is freely available to the public at http://ctrr.bioinf.org.cn/.
- 17 October 2022
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 21 (2) , 292-299
Open AccessCTRR-ncRNA: A Knowledgebase for Cancer Therapy Resistance and Recurrence Associated Non-Coding RNAs
Cancer therapy resistance and recurrence (CTRR) are the dominant causes of death in cancer patients. Recent studies have indicated that non-coding RNAs (ncRNAs) can not only reverse the resistance to cancer therapy but also are crucial biomarkers for the evaluation and prediction of CTRR. Herein, we developed CTRR-ncRNA, a knowledgebase of CTRR-associated ncRNAs, aiming to provide an accurate and comprehensive resource for research involving the association between CTRR and ncRNAs. Compared to most of the existing cancer databases, CTRR-ncRNA is focused on the clinical characterization of cancers, including cancer subtypes, as well as survival outcomes and responses to personalized therapy of cancer patients. Information pertaining to biomarker ncRNAs has also been documented for the development of personalized CTRR prediction. A user-friendly interface and several functional modules have been incorporated into the database. Based on the preliminary analysis of genotype–phenotype relationships, universal ncRNAs have been found to be potential biomarkers for CTRR. The CTRR-ncRNA is a translation-oriented knowledgebase and it provides a valuable resource for mechanistic investigations and explainable artificial intelligence-based modeling. CTRR-ncRNA is freely available to the public at http://ctrr.bioinf.org.cn/.
- Open Access
CTRR-ncRNA: A Translation-oriented Knowledgebase for Cancer Resistance and Recurrence Associated Non-coding RNAs
Cancer therapy resistance and recurrence (CTRR) are the dominant causes of death in cancer patients. Recent studies have indicated that non-coding RNAs (ncRNAs) can not only reverse the resistance to cancer therapy but also are crucial biomarkers for the evaluation and prediction of CTRR. Herein, we developed CTRR-ncRNA, a knowledgebase of CTRR-associated ncRNAs, aiming to provide an accurate and comprehensive resource for research involving association between CTRR and ncRNAs. Compared to most of the existing cancer databases, CTRR-ncRNA is focused on the clinical characterization of cancers, including cancer subtypes, as well as survival outcomes and response to personalized therapy of cancer patients. Information pertaining to biomarker ncRNAs has also been documented for the development of personalized CTRR prediction. A user-friendly interface and several functional modules have been incorporated into the database. Based on the preliminary analysis of genotype–phenotype relationships, universal ncRNAs have been found to be potential biomarkers for CTRR. The CTRR-ncRNA is a translation-oriented knowledgebase and it provides a valuable resource for mechanistic investigations and explainable artificial intelligence-based modelling. CTRR-ncRNA is freely available to the public at http://ctrr.bioinf.org.cn/ .
Open AccessCTRR-ncRNA: A Translation-oriented Knowledgebase for Cancer Resistance and Recurrence Associated Non-coding RNAs
Cancer therapy resistance and recurrence (CTRR) are the dominant causes of death in cancer patients. Recent studies have indicated that non-coding RNAs (ncRNAs) can not only reverse the resistance to cancer therapy but also are crucial biomarkers for the evaluation and prediction of CTRR. Herein, we developed CTRR-ncRNA, a knowledgebase of CTRR-associated ncRNAs, aiming to provide an accurate and comprehensive resource for research involving association between CTRR and ncRNAs. Compared to most of the existing cancer databases, CTRR-ncRNA is focused on the clinical characterization of cancers, including cancer subtypes, as well as survival outcomes and response to personalized therapy of cancer patients. Information pertaining to biomarker ncRNAs has also been documented for the development of personalized CTRR prediction. A user-friendly interface and several functional modules have been incorporated into the database. Based on the preliminary analysis of genotype–phenotype relationships, universal ncRNAs have been found to be potential biomarkers for CTRR. The CTRR-ncRNA is a translation-oriented knowledgebase and it provides a valuable resource for mechanistic investigations and explainable artificial intelligence-based modelling. CTRR-ncRNA is freely available to the public at http://ctrr.bioinf.org.cn/ .