Zhen Li

Dr. Zhen Li


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 January 2026 in Frontiers in Genetics
Verified
Journal Article
Open access
PDF file
SZShulin Zhao
Shulin Zhao
BNBaoyun Nan
Baoyun Nan
JGJun Guo
Jun Guo
WXWenkai Xu
Wenkai Xu
ZLZhen Li
Zhen Li
Introduction: Coronary atherosclerotic heart disease (CHD) is a leading cause of morbidity and mortality worldwide, making timely identification critical for improving patient prognosis. However, traditional imaging examinations are limited by high costs and patient selection bias, while existing prediction models often lack interpretability and generalization ability. This study aimed to develop a robust, interpretable machine learning approach to address these challenges. Methods: This retrospective study analyzed hospitalized patients at Quzhou People’s Hospital from July 2021 to March 2025. Patients diagnosed with CHD were categorized as positive samples, while those without cardiovascular disease served as negative controls. The dataset integrated demographic data, blood biomarkers, and vital signs. A Generative Adversarial Imputation Network (GAIN) was utilized to handle missing values, and multiple machine learning models were constructed and compared for prediction performance. Results: Among the evaluated algorithms, the XGBoost model achieved superior performance on the test set with an Area Under the Curve (AUC) of 0.9053. To enhance clinical utility, the integration of SHAP (SHapley Additive exPlanations) values enabled both global and local interpretation of model decisions. Key predictive factors identified included mean respiratory rate during hospitalization, age, high-sensitivity troponin I (hs-cTnI), and hypertension. Discussion: The developed model demonstrates robust prediction performance combined with high clinical interpretability. Unlike traditional “black box” models, this approach clarifies the contribution of specific risk factors. Crucially, the tool is well-suited for dual deployment: serving as an automated screening tool integrated into hospital electronic health records (EHRs) and as a self-monitoring aid for individuals with underlying health conditions via mobile health applications.

10 November 2025 in Frontiers in Genetics
Verified
Journal Article
Open access
PDF file
YZYixiao Zhai
Yixiao Zhai
ZZZitong Zhang
Zitong Zhang
ZLZhen Li
Zhen Li
The reliability of multiple sequence alignment (MSA) results directly determines the credibility of the conclusions drawn from biological research. However, MSA is inherently an NP-hard problem, making it theoretically impossible to guarantee a globally optimal solution. Consequently, in addition to developing more efficient alignment algorithms, improving the quality of initial alignments through post-processing optimization has become an important strategy for enhancing the overall alignment accuracy. Although post-processing methods have shown potential in improving alignment accuracy, currently, there is a lack of systematic reviews and summaries. In this review, we provide a systematic overview of the development and key research ideas of MSA post-processing methods over the past 3 decades and outline potential directions for future research.

Verified
Journal Article
Open access
PDF file
YZYao Zou
Yao Zou
Quan ZouQuan Zou
Quan Zou
ZLZhen Li
Zhen Li
Colorectal cancer (CRC) is a prevalent and lethal malignancy worldwide. Despite extensive research, core genes for diagnosis and prognosis in CRC remain to be fully elucidated. This study aims to identify novel gene biomarkers for CRC diagnosis and prognosis based on the GEO and TCGA datasets. Integration of TCGA and GEO datasets revealed 197 common differentially expressed genes (DEGs) between CRC tumor and normal samples. Functional enrichment analysis implicated these DEGs in biological processes and signaling pathways critical to CRC progression, including cell cycle regulation and nuclear division. Protein–protein interaction (PPI) network analysis identified 17 hub genes from DEGs, including TROAP, CDKN3, CDCA3, UBE2C, CEP55, KIF11, CDC20, CCNA2, MCM4, CKS2, POLE2, MAD2L1, CCNB1, PTTG1, TPX2, TOP2A, and DLGAP5. All 17 hub genes demonstrated high diagnostic value (AUC > 0.85), including CCNB1 (AUC = 0.944). Based on the Cox proportional hazards regression, an 8-gene prognostic signature (CLCA1, CCNB1, TPM2, MMP3, AOC3, CRYAB, CA4, GUCA2A) effectively stratified patients by survival risk, with a 5-year AUC of 0.71. In vitro, CCNB1 knockdown triggered cell cycle arrest, thereby suppressing the proliferation of colorectal cancer cells. This study validated CCNB1 as a dual-purpose biomarker for CRC diagnosis and favorable prognosis, highlighting its potential utility in clinical management.