Murong Zhou
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
Murong Zhou has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Guohua Wang
College of Computer and Control Engineering, Northeast Forestry University, Harbin, China
6 shared publications - Yuming Zhao
School of Computer Science and Artificial Intelligence, Northeast Forestry University , Harbin 150040
4 shared publications - Qiaoming Liu
College of Artificial Intelligence, Henan University , Zhengzhou 450000
2 shared publications - Runtian Gao
College of Life Sciences, Northeast Forestry University, Harbin, China
2 shared publications Shuqi Cao
Faculty of Computing, Harbin Institute of Technology, Harbin, China
2 shared publications- Tao Jiang
Harbin Institute of Technology
2 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published byĀ Oxford University Press (OUP)Ā inĀ Bioinformatics
- Vol.Ā 42 Ā (11)
TRACER: twin-contrastive learning for fine-grained and rare spatial domain identification in spatial transcriptomics
Motivation: Fine-grained and rare spatial domains are difficult to identify in spatial transcriptomics because they contain few spots, show subtle transcriptional differences, and are strongly influenced by neighboring regions. Results: We present TRACER, a twin-contrastive learning framework that integrates a graph convolutional autoencoder with neighborhood-aware instance-level and cluster-level contrastive learning. By jointly preserving local spatial continuity and enhancing domain separability, TRACER achieved competitive and generally robust performance compared with representative spatial domain identification methods across multiple spatially resolved transcriptomic datasets, including human dorsolateral prefrontal cortex generated by 10Ć Visium, mouse brain datasets generated by STARmap, osmFISH, and BARISTAseq, and human breast cancer. TRACER recovered major cortical and anatomical structures, improved the identification of rare or transitional domains in the mouse somatosensory cortex, and revealed candidate functionally distinct small domains within a broadly annotated breast cancer region. Availability and implementation: The source code of TRACER is available at https://github.com/luxin-heart/TRACER.
- 30 October 2026
- Published byĀ Oxford University Press (OUP)Ā inĀ Bioinformatics
- Vol.Ā 42 Ā (11)
Open AccessTRACER: twin-contrastive learning for fine-grained and rare spatial domain identification in spatial transcriptomics
Motivation: Fine-grained and rare spatial domains are difficult to identify in spatial transcriptomics because they contain few spots, show subtle transcriptional differences, and are strongly influenced by neighboring regions. Results: We present TRACER, a twin-contrastive learning framework that integrates a graph convolutional autoencoder with neighborhood-aware instance-level and cluster-level contrastive learning. By jointly preserving local spatial continuity and enhancing domain separability, TRACER achieved competitive and generally robust performance compared with representative spatial domain identification methods across multiple spatially resolved transcriptomic datasets, including human dorsolateral prefrontal cortex generated by 10Ć Visium, mouse brain datasets generated by STARmap, osmFISH, and BARISTAseq, and human breast cancer. TRACER recovered major cortical and anatomical structures, improved the identification of rare or transitional domains in the mouse somatosensory cortex, and revealed candidate functionally distinct small domains within a broadly annotated breast cancer region. Availability and implementation: The source code of TRACER is available at https://github.com/luxin-heart/TRACER.
- Open Access
- 21 September 2026
- Published byĀ Springer NatureĀ inĀ Nature Methods
- Vol.Ā 23 Ā (10) , 1984-1993
SVPG: a pangenome-based structural variant detection approach and rapid augmentation of pangenome graphs with new samples
Breakthrough advances in long-read sequencing have opened unprecedented opportunities to study genetic variations through pangenome analysis, yet tools that effectively leverage such frameworks for structural variant (SV) detection remain limited. In addition, efficient construction of pangenome graphs becomes increasingly challenging with the acquisition of larger numbers of samples. Here we present SVPG, an approach that leverages haplotype-resolved pangenome reference for accurate SV detection and rapid pangenome graph augmentation from long-read sequencing data. Compared with state-of-the-art SV callers, SVPG maintained superior overall performance across different sequencing technologies and coverages. SVPG also achieved notable improvements in calling individual-specific SVs, including rare and somatic SVs. Furthermore, in a benchmark involving 20 samples, SVPG accelerated pangenome graph augmentation by nearly tenfold compared with traditional augmentation strategies. These results indicate that SVPG has the potential to improve SV detection and serve as an effective tool, offering new possibilities for advancing pangenomic research.
- 21 September 2026
- Published byĀ Springer NatureĀ inĀ Nature Methods
- Vol.Ā 23 Ā (10) , 1984-1993
Open AccessSVPG: a pangenome-based structural variant detection approach and rapid augmentation of pangenome graphs with new samples
Breakthrough advances in long-read sequencing have opened unprecedented opportunities to study genetic variations through pangenome analysis, yet tools that effectively leverage such frameworks for structural variant (SV) detection remain limited. In addition, efficient construction of pangenome graphs becomes increasingly challenging with the acquisition of larger numbers of samples. Here we present SVPG, an approach that leverages haplotype-resolved pangenome reference for accurate SV detection and rapid pangenome graph augmentation from long-read sequencing data. Compared with state-of-the-art SV callers, SVPG maintained superior overall performance across different sequencing technologies and coverages. SVPG also achieved notable improvements in calling individual-specific SVs, including rare and somatic SVs. Furthermore, in a benchmark involving 20 samples, SVPG accelerated pangenome graph augmentation by nearly tenfold compared with traditional augmentation strategies. These results indicate that SVPG has the potential to improve SV detection and serve as an effective tool, offering new possibilities for advancing pangenomic research.
- Open Access
- 1 June 2026
- Published byĀ Oxford University Press (OUP)Ā inĀ Bioinformatics
- Vol.Ā 42 Ā (6)
scTACL: a multitask topology-aware contrastive learning approach for single-cell transcriptomics analysis
Motivation: The advent of single-cell RNA sequencing (scRNA-seq) technology has allowed researchers to measure gene expression profiles at the single-cell level, providing valuable insights into cellular heterogeneity. However, due to the limitations of current sequencing platforms, scRNA-seq data often contain significant noise, particularly severe dropout events, which pose major challenges for subsequent analyses. Results: In this study, we developed a new method called topology-aware contrastive learning (scTACL). This approach uses contrastive learning between a cell similarity graph and a cell embedding similarity graph, employing a zero-inflated negative binomial (ZINB) distribution to model the reconstructed data. This alignment helps the processed data better reflect true biological signals. It delivers superior results in key tasks such as data imputation, clustering, batch effect correction, and cellācell interaction. Additionally, scTACL successfully identified two distinct subtypes of epithelial cells in lung adenocarcinoma tissues, further demonstrating its effectiveness and usefulness in complex biological settings. Notably, without relying on spatial location information, scTACL still effectively distinguished the epithelial and mesenchymal regions in the spatial transcriptome data of liver cancer and identified the COLLAGEN signaling pathway, which plays a crucial role in the epithelialāmesenchymal transition process through intercellular communication analysis.
- 1 June 2026
- Published byĀ Oxford University Press (OUP)Ā inĀ Bioinformatics
- Vol.Ā 42 Ā (6)
Open AccessscTACL: a multitask topology-aware contrastive learning approach for single-cell transcriptomics analysis
Motivation: The advent of single-cell RNA sequencing (scRNA-seq) technology has allowed researchers to measure gene expression profiles at the single-cell level, providing valuable insights into cellular heterogeneity. However, due to the limitations of current sequencing platforms, scRNA-seq data often contain significant noise, particularly severe dropout events, which pose major challenges for subsequent analyses. Results: In this study, we developed a new method called topology-aware contrastive learning (scTACL). This approach uses contrastive learning between a cell similarity graph and a cell embedding similarity graph, employing a zero-inflated negative binomial (ZINB) distribution to model the reconstructed data. This alignment helps the processed data better reflect true biological signals. It delivers superior results in key tasks such as data imputation, clustering, batch effect correction, and cellācell interaction. Additionally, scTACL successfully identified two distinct subtypes of epithelial cells in lung adenocarcinoma tissues, further demonstrating its effectiveness and usefulness in complex biological settings. Notably, without relying on spatial location information, scTACL still effectively distinguished the epithelial and mesenchymal regions in the spatial transcriptome data of liver cancer and identified the COLLAGEN signaling pathway, which plays a crucial role in the epithelialāmesenchymal transition process through intercellular communication analysis.
- Open Access
Cost-effective hybrid long- and short-read sequencing enables accurate somatic structural variant detection
Somatic structural variant (SSV) calling typically requires matched normal data. Incorporating relatively inexpensive short-read sequencing not only provides essential germline information but can also replace a substantial portion of long-read sequencing, thereby enabling more cost-effective somatic SV detection. Here, we present SomaSV, a hybrid sequencing framework that integrates 30Ć tumor long-read data with matched normal data comprising 10Ć long-read and 30Ć short-read sequencing. This design achieves high-accuracy somatic SV detection while remaining cost-competitive. Comprehensive benchmarking demonstrates that SomaSV outperforms current state-of-the-art methods by more than 13% in F1 score while reducing sequencing costs by approximately 19%. Moreover, SomaSV identifies clinically relevant cancer-associated genes, including CLDN4 and ROBO2 , highlighting its potential for discovering valuable biomarkers to support early cancer screening and diagnosis. The source code for SomaSV can be accessed at https://github.com/eioyuou/SomaSV .
Open AccessCost-effective hybrid long- and short-read sequencing enables accurate somatic structural variant detection
Somatic structural variant (SSV) calling typically requires matched normal data. Incorporating relatively inexpensive short-read sequencing not only provides essential germline information but can also replace a substantial portion of long-read sequencing, thereby enabling more cost-effective somatic SV detection. Here, we present SomaSV, a hybrid sequencing framework that integrates 30Ć tumor long-read data with matched normal data comprising 10Ć long-read and 30Ć short-read sequencing. This design achieves high-accuracy somatic SV detection while remaining cost-competitive. Comprehensive benchmarking demonstrates that SomaSV outperforms current state-of-the-art methods by more than 13% in F1 score while reducing sequencing costs by approximately 19%. Moreover, SomaSV identifies clinically relevant cancer-associated genes, including CLDN4 and ROBO2 , highlighting its potential for discovering valuable biomarkers to support early cancer screening and diagnosis. The source code for SomaSV can be accessed at https://github.com/eioyuou/SomaSV .
- Open Access
- 1 November 2025
- Published byĀ Oxford University Press (OUP)Ā inĀ Briefings in Bioinformatics
- Vol.Ā 26 Ā (6)
SLGCA: spatial cross-level graph contrastive autoencoder for multislice spatial domain identification and microenvironment exploration
The development of spatial transcriptomics (ST) technologies has enabled researchers to better understand cellsā spatial organization and functional heterogeneity within their native tissue context. Spatial domain identification plays a crucial role in ST data analysis. However, most existing spatial domain identification methods do not fully exploit spatial information, and often fail to adequately integrate both local and global features, resulting in suboptimal spatial domain identification. We propose SLGCA, a novel method based on cross-level graph contrastive learning to address these challenges. SLGCA adopts a dual-channel learning mechanism, combining local-level contrastive learning based on spatial neighborhood information and global information contrastive learning across views, thereby significantly enhancing the accuracy of spatial domain identification. SLGCA can integrate multiple tissue sections without needing pre-alignment or external tools, eliminating batch effects and accurately identifying spatial domains across multiple slices. Experimental results show that SLGCA significantly outperforms the benchmark methods in spatial domain identification accuracy on ST data generated by multiple techniques. Moreover, SLGCA enables accurate dissection of tumor heterogeneity in human breast cancer datasets and effectively uncovers the heterogeneous tumor microenvironment in liver cancer, revealing two distinct fibroblast subtypes.
- 1 November 2025
- Published byĀ Oxford University Press (OUP)Ā inĀ Briefings in Bioinformatics
- Vol.Ā 26 Ā (6)
Open AccessSLGCA: spatial cross-level graph contrastive autoencoder for multislice spatial domain identification and microenvironment exploration
The development of spatial transcriptomics (ST) technologies has enabled researchers to better understand cellsā spatial organization and functional heterogeneity within their native tissue context. Spatial domain identification plays a crucial role in ST data analysis. However, most existing spatial domain identification methods do not fully exploit spatial information, and often fail to adequately integrate both local and global features, resulting in suboptimal spatial domain identification. We propose SLGCA, a novel method based on cross-level graph contrastive learning to address these challenges. SLGCA adopts a dual-channel learning mechanism, combining local-level contrastive learning based on spatial neighborhood information and global information contrastive learning across views, thereby significantly enhancing the accuracy of spatial domain identification. SLGCA can integrate multiple tissue sections without needing pre-alignment or external tools, eliminating batch effects and accurately identifying spatial domains across multiple slices. Experimental results show that SLGCA significantly outperforms the benchmark methods in spatial domain identification accuracy on ST data generated by multiple techniques. Moreover, SLGCA enables accurate dissection of tumor heterogeneity in human breast cancer datasets and effectively uncovers the heterogeneous tumor microenvironment in liver cancer, revealing two distinct fibroblast subtypes.
- Open Access
- 1 May 2025
- Published byĀ Oxford University Press (OUP)Ā inĀ Briefings in Bioinformatics
- Vol.Ā 26 Ā (3)
Enhancing LncRNA-miRNA interaction prediction with multimodal contrastive representation learning
Interactions between long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) play an important role in the development of complex human diseases by collaboratively regulating gene transcription and expression. Therefore, identifying lncRNA-miRNA interactions (LMIs) is essential for diagnosing and treating complex human diseases. Because identifying LMIs with wet experiments is time-consuming and labor-intensive, some computational methods have been developed to infer LMIs. However, these approaches excel at utilizing single-modal information but struggle to integrate multimodal data from lncRNAs and miRNAs, which is essential for uncovering complex patterns in LMIs, ultimately limiting their performance. Therefore, this article proposes a novel multimodal contrastive representation learning model (MCRLMI) for LMI predictions. The model fully integrates multi-source similarity information and sequence encodings of lncRNAs and miRNAs. It leverages a graph convolutional network (GCN) and a Transformer to capture local neighborhood structural features and long-distance dependencies, respectively, enabling the collaborative modeling of structural and semantic information. Subsequently, to effectively integrate multimodal characteristics with encoded information, a multichannel attention mechanism and contrastive learning are introduced to fuse the extracted features. Finally, a KolmogorovāArnold Network (KAN) is trained with the optimized embeddings to predict LMIs. Extensive experiments show that the proposed MCRLMI consistently outperforms existing methods. Moreover, case studies further validate the potential of MCRLMI to identify novel LMIs in practical applications.
- 1 May 2025
- Published byĀ Oxford University Press (OUP)Ā inĀ Briefings in Bioinformatics
- Vol.Ā 26 Ā (3)
Open AccessEnhancing LncRNA-miRNA interaction prediction with multimodal contrastive representation learning
Interactions between long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) play an important role in the development of complex human diseases by collaboratively regulating gene transcription and expression. Therefore, identifying lncRNA-miRNA interactions (LMIs) is essential for diagnosing and treating complex human diseases. Because identifying LMIs with wet experiments is time-consuming and labor-intensive, some computational methods have been developed to infer LMIs. However, these approaches excel at utilizing single-modal information but struggle to integrate multimodal data from lncRNAs and miRNAs, which is essential for uncovering complex patterns in LMIs, ultimately limiting their performance. Therefore, this article proposes a novel multimodal contrastive representation learning model (MCRLMI) for LMI predictions. The model fully integrates multi-source similarity information and sequence encodings of lncRNAs and miRNAs. It leverages a graph convolutional network (GCN) and a Transformer to capture local neighborhood structural features and long-distance dependencies, respectively, enabling the collaborative modeling of structural and semantic information. Subsequently, to effectively integrate multimodal characteristics with encoded information, a multichannel attention mechanism and contrastive learning are introduced to fuse the extracted features. Finally, a KolmogorovāArnold Network (KAN) is trained with the optimized embeddings to predict LMIs. Extensive experiments show that the proposed MCRLMI consistently outperforms existing methods. Moreover, case studies further validate the potential of MCRLMI to identify novel LMIs in practical applications.
- 30 December 2024
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Fuzzy Systems
- Vol.Ā 33 Ā (5) , 1429-1440
Adjacency-Aware Fuzzy Label Learning for Skin Disease Diagnosis
Automatic acne severity grading is crucial for the accurate diagnosis and effective treatment of skin diseases. However, the acne severity grading process is often ambiguous due to the similar appearance of acne with close severity, making it challenging to achieve reliable acne severity grading. Following the idea of fuzzy logic for handling uncertainty in decision-making, we transforms the acne severity grading task into a fuzzy label learning problem, and propose a novel Adjacency-aware Fuzzy Label Learning (AFLL) framework to handle uncertainties in this task. The AFLL framework makes four significant contributions, each demonstrated to be highly effective in extensive experiments. First, we introduce a novel adjacency-aware decision sequence generation method that enhances sequence tree construction by reducing bias and improving discriminative power. Second, we present a consistency-guided decision sequence prediction method that mitigates error propagation in hierarchical decision-making through a novel selective masking decision strategy. Third, our proposed sequential conjoint distribution loss innovatively captures the differences for both high and low fuzzy memberships across the entire fuzzy label set while modeling the internal temporal order among different acne severity labels with a cumulative distribution, leading to substantial improvements in fuzzy label learning. Fourth, to the best of our knowledge, AFLL is the first approach to explicitly address the challenge of distinguishing adjacent categories in acne severity grading tasks. Experimental results on the public ACNE04 dataset demonstrate that AFLL significantly outperforms existing methods, establishing a new state-of-the-art in acne severity grading.
- 30 December 2024
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Fuzzy Systems
- Vol.Ā 33 Ā (5) , 1429-1440
Adjacency-Aware Fuzzy Label Learning for Skin Disease Diagnosis
Automatic acne severity grading is crucial for the accurate diagnosis and effective treatment of skin diseases. However, the acne severity grading process is often ambiguous due to the similar appearance of acne with close severity, making it challenging to achieve reliable acne severity grading. Following the idea of fuzzy logic for handling uncertainty in decision-making, we transforms the acne severity grading task into a fuzzy label learning problem, and propose a novel Adjacency-aware Fuzzy Label Learning (AFLL) framework to handle uncertainties in this task. The AFLL framework makes four significant contributions, each demonstrated to be highly effective in extensive experiments. First, we introduce a novel adjacency-aware decision sequence generation method that enhances sequence tree construction by reducing bias and improving discriminative power. Second, we present a consistency-guided decision sequence prediction method that mitigates error propagation in hierarchical decision-making through a novel selective masking decision strategy. Third, our proposed sequential conjoint distribution loss innovatively captures the differences for both high and low fuzzy memberships across the entire fuzzy label set while modeling the internal temporal order among different acne severity labels with a cumulative distribution, leading to substantial improvements in fuzzy label learning. Fourth, to the best of our knowledge, AFLL is the first approach to explicitly address the challenge of distinguishing adjacent categories in acne severity grading tasks. Experimental results on the public ACNE04 dataset demonstrate that AFLL significantly outperforms existing methods, establishing a new state-of-the-art in acne severity grading.
- 1 January 2022
- Published byĀ Bentham Science Publishers Ltd.Ā inĀ Current Bioinformatics
- Vol.Ā 17 Ā (1) , 31-36
A Review of DNA Data Storage Technologies Based on Biomolecules
: In the information age, data storage technology has become the key to improving computer systems. Since traditional storage technologies cannot meet the demand for massive storage, new DNA storage technology based on biomolecules attracts much attention. DNA storage refers to the technology that uses artificially synthesized deoxynucleotide chains to store and read all information, such as documents, pictures, and audio. First, data are encoded into binary number strings. Then, the four types of base, A(Adenine), T(Thymine), C(Cytosine), and G(Guanine), are used to encode the corresponding binary numbers so that the data can be used to construct the target DNA molecules in the form of deoxynucleotide chains. Subsequently, the corresponding DNA molecules are artificially synthesized, enabling the data to be stored within them. Compared with traditional storage systems, DNA storage has major advantages, such as high storage density, long duration, as well as low hardware cost, high access parallelism, and strong scalability, which satisfies the demands for big data storage. This manuscript first reviews the origin and development of DNA storage technology, then the storage principles, contents, and methods are introduced. Finally, the development of DNA storage technology is analyzed. From the initial research to the cutting edge of this field and beyond, the advantages, disadvantages, and practical applications of DNA storage technology require continuous exploration.
- 1 January 2022
- Published byĀ Bentham Science Publishers Ltd.Ā inĀ Current Bioinformatics
- Vol.Ā 17 Ā (1) , 31-36
A Review of DNA Data Storage Technologies Based on Biomolecules
: In the information age, data storage technology has become the key to improving computer systems. Since traditional storage technologies cannot meet the demand for massive storage, new DNA storage technology based on biomolecules attracts much attention. DNA storage refers to the technology that uses artificially synthesized deoxynucleotide chains to store and read all information, such as documents, pictures, and audio. First, data are encoded into binary number strings. Then, the four types of base, A(Adenine), T(Thymine), C(Cytosine), and G(Guanine), are used to encode the corresponding binary numbers so that the data can be used to construct the target DNA molecules in the form of deoxynucleotide chains. Subsequently, the corresponding DNA molecules are artificially synthesized, enabling the data to be stored within them. Compared with traditional storage systems, DNA storage has major advantages, such as high storage density, long duration, as well as low hardware cost, high access parallelism, and strong scalability, which satisfies the demands for big data storage. This manuscript first reviews the origin and development of DNA storage technology, then the storage principles, contents, and methods are introduced. Finally, the development of DNA storage technology is analyzed. From the initial research to the cutting edge of this field and beyond, the advantages, disadvantages, and practical applications of DNA storage technology require continuous exploration.
- Open Access
- 1 January 2021
- Published byĀ WileyĀ inĀ BioMed Research International
- Vol.Ā 2021 Ā (1)
LCāMS/MSāBased Quantitative Proteomics Analysis of Different Stages of NonāSmallāCell Lung Cancer
Lung cancer has a higher incidence rate and mortality rate than all other cancers. Early diagnosis and treatment of lung cancer remain a major challenge, and the 5āyear survival rate of its patients is only 15%. Basic and clinical research, especially the discovery of biomarkers, is crucial for improving the diagnosis and treatment of lung cancer patients. To identify novel biomarkers for lung cancer, we used the iTRAQ8āplex labeling technology combined with liquid chromatographyātandem mass spectrometry (LCāMS/MS) to analyze the serum and urine of patients with different stages of lung adenocarcinoma and healthy individuals. A total of 441 proteins were identified in the serum, and 1,161 proteins were identified in the urine. The levels of elongation factor 1āalpha 2, proteasome subunit alpha type, and spermatogenesisāassociated protein increased significantly in the serum of patients with lung cancer compared with those in healthy controls. The levels of transmembrane protein 143, cadherin 5, fibronectin 1, and collectinā11 decreased significantly in the serum of patients with metastases compared with those of nonmetastatic lung cancer patients. In the urine of stage III and IV lung cancer patients, the prostateāspecific antigen and prostatic acid phosphatase decreased significantly, whereas neutrophil defensin 1 increased significantly. The results of LCāMS/MS were confirmed by enzymeālinked immunosorbent assay (ELISA) for transmembrane protein 143, cadherin 5, fibronectin 1, and collectinā11 in the serum. These proteins may be a potential early diagnosis and metastasis biomarkers for lung adenocarcinoma. Furthermore, the relative content of these markers in the serum and urine could be used to determine the progression of lung adenocarcinoma and achieve accurate staging and diagnosis.
- 1 January 2021
- Published byĀ WileyĀ inĀ BioMed Research International
- Vol.Ā 2021 Ā (1)
Open AccessLCāMS/MSāBased Quantitative Proteomics Analysis of Different Stages of NonāSmallāCell Lung Cancer
Lung cancer has a higher incidence rate and mortality rate than all other cancers. Early diagnosis and treatment of lung cancer remain a major challenge, and the 5āyear survival rate of its patients is only 15%. Basic and clinical research, especially the discovery of biomarkers, is crucial for improving the diagnosis and treatment of lung cancer patients. To identify novel biomarkers for lung cancer, we used the iTRAQ8āplex labeling technology combined with liquid chromatographyātandem mass spectrometry (LCāMS/MS) to analyze the serum and urine of patients with different stages of lung adenocarcinoma and healthy individuals. A total of 441 proteins were identified in the serum, and 1,161 proteins were identified in the urine. The levels of elongation factor 1āalpha 2, proteasome subunit alpha type, and spermatogenesisāassociated protein increased significantly in the serum of patients with lung cancer compared with those in healthy controls. The levels of transmembrane protein 143, cadherin 5, fibronectin 1, and collectinā11 decreased significantly in the serum of patients with metastases compared with those of nonmetastatic lung cancer patients. In the urine of stage III and IV lung cancer patients, the prostateāspecific antigen and prostatic acid phosphatase decreased significantly, whereas neutrophil defensin 1 increased significantly. The results of LCāMS/MS were confirmed by enzymeālinked immunosorbent assay (ELISA) for transmembrane protein 143, cadherin 5, fibronectin 1, and collectinā11 in the serum. These proteins may be a potential early diagnosis and metastasis biomarkers for lung adenocarcinoma. Furthermore, the relative content of these markers in the serum and urine could be used to determine the progression of lung adenocarcinoma and achieve accurate staging and diagnosis.