Identification Of Cancer Subtypes Research Articles

PCA-constrained multi-core matrix fusion network: A Novel Approach for Cancer Subtype Identification

PCA-constrained multi-core matrix fusion network: A Novel Approach for Cancer Subtype Identification

Multi-kernel subspace stable clustering with exact rank constraints

The identification of cancer subtypes plays a critical role in the early diagnosis of cancer and the delivery of appropriate treatment. Clustering based on comprehensive multi-omic data often yields superior results than clustering using a single data type since each omic type may contain complementary information. However, existing clustering methods cannot fully explore the internal structural information within multi-omic data and can lead to unstable clustering results. This paper proposes a novel multi-omic clustering framework, Multi-Kernel Subspace Stable Clustering with Exact Rank Constraints (MKSSC-ERC), which can effectively explore the correlation and complementary information between different omics types. Specifically, we design a kernel selection criterion that considers both accuracy and diversity and each omic data gets a base kernel. Then, we extract a consistent affinity matrix for these base kernels using subspace segmentation with exact rank constraints, which can refine the extracted shared structures and make them robust to sample noises. In addition, we devise a strategy for selecting the optimal initial cluster centers, which significantly enhances the stability of clustering results. Simulation studies on benchmark multi-omic datasets illustrate substantial gains over existing state-of-the-art methods in survival analysis. MKSSC-ERC is available at https://github.com/xuzihan66/MKSSC-ERC.

Cervical cancer subtype identification and model building based on lipid metabolism and post-infection microenvironment immune landscape

BackgroundAs the second most common gynecological cancer, cervical cancer (CC) seriously threatens women's health. The poor prognosis of CC is closely related to the post-infection microenvironment (PIM). This study investigated how lipid metabolism-related genes (LMRGs) affect CC PIM and their role in diagnosing CC. MethodsWe analyzed lipid metabolism scores in the CC single-cell landscape by AUCell. The differentiation trajectory of epithelial cells to cancer cells was revealed using LMRGs and Monocle2. Consensus clustering was used to identify novel subgroups using the LMRGs. Multiple immune assessment methods were used to evaluate the immune landscape of the subgroups. Prognostic genes were determined by the LASSO and multivariate Cox regression analysis. Finally, we perform molecular docking of prognostic genes to explore potential therapeutic agents. ResultsWe revealed the differentiation trajectory of epithelial cells to cancer cells in CC by LMRGs. The higher LMRGs expression cluster had higher survival rates and immune infiltration expression. Functional enrichment showed that two clusters were mainly involved in immune response regulation. A novel LMR signature (LMR.sig) was constructed to predict clinical outcomes in CC. The expression of prognostic genes was correlated with the PIM immune landscape. Small molecular compounds with the best binding effect to prognostic genes were obtained by molecular docking, which may be used as new targeted therapeutic drugs. ConclusionWe found that the subtype with better prognosis could regulate the expression of some critical genes through more frequent lipid metabolic reprogramming, thus affecting the maturation and migration of dendritic cells (DCs) and the expression of M1 macrophages, reshaping the immunosuppressive environment of PIM in CC patients. LMRGs are closely related to the PIM immune landscape and can accurately predict tumor prognosis. These results further our understanding of the underlying mechanisms of LMRGs in CC.

Abstract 7566: Identification of cancer subtypes with a ctDNA-based targeted methylation assay

Abstract Introduction: Identifying cancer subtypes is necessary for cancer diagnosis, prognosis determination, and treatment selection. Furthermore, as transformation between subtypes is increasingly recognized as a key resistance mechanism to targeted therapies, serial subtype reassessment is likely to gain adoption. Cancer subtypes have traditionally been determined by pathologists based on histology; more recently, molecular subtyping has been performed using IHC, RNAseq, or assays that detect genetic alterations. However, these methods have significant limitations, including tissue biopsy requirement (IHC and RNAseq), low discriminative resolution for distinct transcriptional programs (IHC and mutation detection), and poor reproducibility and feasibility (RNAseq). The GRAIL plasma-only, ctDNA-based targeted methylation platform is a robust, biopsy-free, scalable assay that can distinguish cancer methylation patterns between different cancer signal origins through the use of a proprietary classifier. Here, we explore the GRAIL platform’s potential to detect finer-scale differences in cancer biology by developing classification algorithms for subtyping of three common cancer types. Methods: As part of the Circulating Cell-free Genome Atlas (NCT02889978) and STRIVE (NCT03085888) studies, clinical data were recorded and plasma samples were collected, accessioned, stored, and processed through GRAIL’s targeted methylation assay from 3,989 cancer and 6,013 non-cancer participants. We developed algorithms to predict cancer status and subtype for breast (triple negative breast cancer [TNBC] model; 531 cancer training participants), lung (lung histology model; 334 cancer training participants), and head and neck (HPV model; 3090 cancer training participants) cancers. Dimensionality reduction of methylation data was performed to illustrate subtype separation within these cancer types. Performance of these three algorithms was assessed on a held-out cohort of cancer samples (184 breast, 241 lung, 67 head and neck) with detectable ctDNA. Results: The lung histology model correctly classified 95% (105/111) of adenocarcinomas, 88% (68/77) of squamous cell carcinomas, and 94% (59/63) of lung neuroendocrine carcinomas and tumors. The TNBC model correctly classified 84% (58/69) of TNBCs and 82% (94/115) of non-TNBCs. The HPV model correctly identified 98% (48/49) of HPV-positive head and neck cancers, and 89% (16/18) of HPV-negative head and neck cancers. Conclusions: GRAIL’s proprietary ctDNA-based targeted methylation platform demonstrates the ability to accurately predict cancer subtypes in multiple cancers using only a plasma sample without the need for tumor biopsy. Future efforts will be directed towards generalizing the GRAIL subtyping method to additional cohorts and cancer types. Citation Format: Tracy Nance, Timothy Shaver, Yifan Zhou, Margaret Antonio, Yinjie Gao, Jonathan Heiss, Robert Calef, Joseph Hiatt, Oliver Venn, Joerg Bredno, John Beausang, Lisa Newman, Charles Swanton, Chun Zhang. Identification of cancer subtypes with a ctDNA-based targeted methylation assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7566.

Deep learning-based identification of esophageal cancer subtypes through analysis of high-resolution histopathology images.

Esophageal cancer (EC) remains a significant health challenge globally, with increasing incidence and high mortality rates. Despite advances in treatment, there remains a need for improved diagnostic methods and understanding of disease progression. This study addresses the significant challenges in the automatic classification of EC, particularly in distinguishing its primary subtypes: adenocarcinoma and squamous cell carcinoma, using histopathology images. Traditional histopathological diagnosis, while being the gold standard, is subject to subjectivity and human error and imposes a substantial burden on pathologists. This study proposes a binary class classification system for detecting EC subtypes in response to these challenges. The system leverages deep learning techniques and tissue-level labels for enhanced accuracy. We utilized 59 high-resolution histopathological images from The Cancer Genome Atlas (TCGA) Esophageal Carcinoma dataset (TCGA-ESCA). These images were preprocessed, segmented into patches, and analyzed using a pre-trained ResNet101 model for feature extraction. For classification, we employed five machine learning classifiers: Support Vector Classifier (SVC), Logistic Regression (LR), Decision Tree (DT), AdaBoost (AD), Random Forest (RF), and a Feed-Forward Neural Network (FFNN). The classifiers were evaluated based on their prediction accuracy on the test dataset, yielding results of 0.88 (SVC and LR), 0.64 (DT and AD), 0.82 (RF), and 0.94 (FFNN). Notably, the FFNN classifier achieved the highest Area Under the Curve (AUC) score of 0.92, indicating its superior performance, followed closely by SVC and LR, with a score of 0.87. This suggested approach holds promising potential as a decision-support tool for pathologists, particularly in regions with limited resources and expertise. The timely and precise detection of EC subtypes through this system can substantially enhance the likelihood of successful treatment, ultimately leading to reduced mortality rates in patients with this aggressive cancer.

Identification of breast cancer subtypes based on graph convolutional network

Identification of molecular subtypes of malignant tumors plays a vital role in individualized diagnosis, personalized treatment, and prognosis prediction of cancer patients. The continuous improvement of comprehensive tumor genomics database and the ongoing breakthroughs in deep learning technology have driven further advancements in computer-aided tumor classification. Although the existing classification methods based on gene expression omnibus database take the complexity of cancer molecular classification into account, they ignore the internal correlation and synergism of genes. To solve this problem, we propose a multi-layer graph convolutional network model for breast cancer subtype classification combined with hierarchical attention network. This model constructs the graph embedding datasets of patients' genes, and develops a new end-to-end multi-classification model, which can effectively recognize molecular subtypes of breast cancer. A large number of test data prove the good performance of this new model in the classification of breast cancer subtypes. Compared to the original graph convolutional neural networks and two mainstream graph neural network classification algorithms, the new model has remarkable advantages. The accuracy, weight-F1-score, weight-recall, and weight-precision of our model in seven-category classification has reached 0.851 7, 0.823 5, 0.851 7 and 0.793 6 respectively. In the four-category classification, the results are 0.928 5, 0.894 9, 0.928 5 and 0.865 0 respectively. In addition, compared with the latest breast cancer subtype classification algorithms, the method proposed in this paper also achieved the highest classification accuracy. In summary, the model proposed in this paper may serve as an auxiliary diagnostic technology, providing a reliable option for precise classification of breast cancer subtypes in the future and laying the theoretical foundation for computer-aided tumor classification.

Supervised graph contrastive learning for cancer subtype identification through multi-omics data integration.

Cancer is one of the most deadly diseases in the world. Accurate cancer subtype classification is critical for patient diagnosis, treatment, and prognosis. Ever-increasing multi-omics data describes the characteristics of the patients from different views and serves as complementary information to promote cancer subtype identification. However, omics data generally have different distributions and high dimensions. How to effectively integrate multiple omics data to classify cancer subtypes accurately is a challenge for researchers. This work proposes a method integrating multi-omics data based on supervised graph contrast learning (MCRGCN) to classify cancer subtypes. The method considers the unique feature distribution of each omics data and the interaction of different omics data features to improve the accuracy of cancer subtype classification. To achieve this, MCRGCN first constructs different sample networks based on the multi-omics data of the samples. Then, it puts the omics data and adjacency matrix of the sample into different residual graph convolution models to get multi-omics features of the samples, which are trained with a supervised comparison loss to maintain that the sample features of each omics should be as consistent as possible. Finally, we input the sample features combining multi-omics features into a classifier to obtain the cancer subtypes. We applied MCRGCN to the invasive breast carcinoma (BRCA) and glioblastoma multiforme (GBM) datasets, integrating gene expression, miRNA expression, and DNA methylation data. The results demonstrate that our model is superior to other methods in integrating multi-omics data. Moreover, the results of survival analysis experiments demonstrate that the cancer subtypes identified by our model have significant clinical features. Furthermore, our model can help to identify potential biomarkers and pathways associated with cancer subtypes.

WMKL: multi-omics data integration enables novel cancer subtype identification via weight-boosted multi-kernel learning.

Cancer is a heterogeneous disease driven by complex molecular alterations. Cancer subtypes determined from multi-omics data can provide novel insight into personalised precision treatment. It is recognised that incorporating prior weight knowledge into multi-omics data integration can improve disease subtyping. We develop a weighted method, termed weight-boosted Multi-Kernel Learning (wMKL) which incorporates heterogeneous data types as well as flexible weight functions, to boost subtype identification. Given a series of weight functions, we propose an omnibus combination strategy to integrate different weight-related P-values to improve subtyping precision. wMKL models each data type with multiple kernel choices, thus alleviating the sensitivity and robustness issue due to selecting kernel parameters. Furthermore, wMKL integrates different data types by learning weights of different kernels derived from each data type, recognising the heterogeneous contribution of different data types to the final subtyping performance. The proposed wMKL outperforms existing weighted and non-weighted methods. The utility and advantage of wMKL are illustrated through extensive simulations and applications to two TCGA datasets. Novel subtypes are identified followed by extensive downstream bioinformatics analysis to understand the molecular mechanisms differentiating different subtypes. The proposed wMKL method provides a novel strategy for disease subtyping. The wMKL is freely available at https://github.com/biostatcao/wMKL .

Identification of colon cancer subtypes based on multi-omics data-construction of methylation markers for immunotherapy.

Being the most widely used biomarker for immunotherapy, the microsatellite status has limitations in identifying all patients who benefit in clinical practice. It is essential to identify additional biomarkers to guide immunotherapy. Aberrant DNA methylation is consistently associated with changes in the anti-tumor immune response, which can promote tumor progression. This study aims to explore immunotherapy biomarkers for colon cancers from the perspective of DNA methylation. The related data (RNA sequencing data and DNA methylation data) were obtained from The Cancer Genome Atlas (TCGA) and UCSC XENA database. Methylation-driven genes (MDGs) were identified through the Pearson correlation analysis. Unsupervised consensus clustering was conducted using these MDGs to identify distinct clusters of colon cancers. Subsequently, we evaluated the immune status and predicted the efficacy of immunotherapy by tumor immune dysfunction and exclusion (Tide) score. Finally, The Quantitative Differentially Methylated Regions (QDMR) software was used to identify the specific DNA methylation markers within particular clusters. A total of 282 MDGs were identified by integrating the DNA methylation and RNA-seq data. Consensus clustering using the K-means algorithm revealed that the optimal number of clusters was 4. It was revealed that the composition of the tumor immune microenvironment (TIME) in Cluster 1 was significantly different from others, and it exhibited a higher level of tumor mutation burdens (TMB) and stronger anti-tumor immune activity. Furthermore, we identified three specific hypermethylation genes that defined Cluster 1 (PCDH20, APCDD1, COCH). Receiver operating characteristic (ROC) curves demonstrated that these specific markers could effectively distinguish Cluster 1 from other clusters, with an AUC of 0.947 (95% CI 0.903-0.990). Finally, we selected clinical samples for immunohistochemical validation. In conclusion, through the analysis of DNA methylation, consensus clustering of colon cancer could effectively identify the cluster that benefit from immunotherapy along with specific methylation biomarkers.

Multi-omics integration with weighted affinity and self-diffusion applied for cancer subtypes identification

BackgroundCharacterizing cancer molecular subtypes is crucial for improving prognosis and individualized treatment. Integrative analysis of multi-omics data has become an important approach for disease subtyping, yielding better understanding of the complex biology. Current multi-omics integration tools and methods for cancer subtyping often suffer challenges of high computational efficiency as well as the problem of weight assignment on data types.ResultsHere, we present an efficient multi-omics integration via weighted affinity and self-diffusion (MOSD) to dissect cancer heterogeneity. MOSD first construct local scaling affinity on each data type and then integrate all affinities by weighted linear combination, followed by the self-diffusion to further improve the patients’ similarities for the downstream clustering analysis. To demonstrate the effectiveness and usefulness for cancer subtyping, we apply MOSD across ten cancer types with three measurements (Gene expression, DNA methylation, miRNA).ConclusionsOur approach exhibits more significant differences in patient survival and computationally efficient benchmarking against several state-of-art integration methods and the identified molecular subtypes reveal strongly biological interpretability. The code as well as its implementation are available in GitHub: https://github.com/DXCODEE/MOSD.

Cancer Subtype Identification Through Integrating Inter and Intra Dataset Relationships in Multi-Omics Data

Cancer Subtype Identification Through Integrating Inter and Intra Dataset Relationships in Multi-Omics Data

Discriminative Deep Canonical Correlation Analysis for Multi-View Data.

Over the past few years, multimodal data analysis has emerged as an inevitable method for identifying sample categories. In the multi-view data classification problem, it is expected that the joint representation should include the supervised information of sample categories so that the similarity in the latent space implies the similarity in the corresponding concepts. Since each view has different statistical properties, the joint representation should be able to encapsulate the underlying nonlinear data distribution of the given observations. Another important aspect is the coherent knowledge of the multiple views. It is required that the learning objective of the multi-view model efficiently captures the nonlinear correlated structures across different modalities. In this context, this article introduces a novel architecture, termed discriminative deep canonical correlation analysis (D2CCA), for classifying given observations into multiple categories. The learning objective of the proposed architecture includes the merits of generative models to identify the underlying probability distribution of the given observations. In order to improve the discriminative ability of the proposed architecture, the supervised information is incorporated into the learning objective of the proposed model. It also enables the architecture to serve as both a feature extractor as well as a classifier. The theory of CCA is integrated with the objective function so that the joint representation of the multi-view data is learned from maximally correlated subspaces. The proposed framework is consolidated with corresponding convergence analysis. The efficacy of the proposed architecture is studied on different domains of applications, namely, object recognition, document classification, multilingual categorization, face recognition, and cancer subtype identification with reference to several state-of-the-art methods.

Identification of Breast Cancer Subtypes Based on Endoplasmic Reticulum Stress-Related Genes and Analysis of Prognosis and Immune Microenvironment in Breast Cancer Patients.

Introduction: Endoplasmic reticulum stress (ERS) was a response to the accumulation of unfolded proteins and plays a crucial role in the development of tumors, including processes such as tumor cell invasion, metastasis, and immune evasion. However, the specific regulatory mechanisms of ERS in breast cancer (BC) remain unclear. Methods: In this study, we analyzed RNA sequencing data from The Cancer Genome Atlas (TCGA) for breast cancer and identified 8 core genes associated with ERS: ELOVL2, IFNG, MAP2K6, MZB1, PCSK6, PCSK9, IGF2BP1, and POP1. We evaluated their individual expression, independent diagnostic, and prognostic values in breast cancer patients. A multifactorial Cox analysis established a risk prognostic model, validated with an external dataset. Additionally, we conducted a comprehensive assessment of immune infiltration and drug sensitivity for these genes. Results: The results indicate that these eight core genes play a crucial role in regulating the immune microenvironment of breast cancer (BRCA) patients. Meanwhile, an independent diagnostic model based on the expression of these eight genes shows limited independent diagnostic value, and its independent prognostic value is unsatisfactory, with the time ROC AUC values generally below 0.5. According to the results of logistic regression neural networks and risk prognosis models, when these eight genes interact synergistically, they can serve as excellent biomarkers for the diagnosis and prognosis of breast cancer patients. Furthermore, the research findings have been confirmed through qPCR experiments and validation. Conclusion: In conclusion, we explored the mechanisms of ERS in BRCA patients and identified 8 outstanding biomolecular diagnostic markers and prognostic indicators. The research results were double-validated using the GEO database and qPCR.

Joint Learning of Spectral Clustering and Low-Rank Representation Based on Precise Segmentation Cost for Cancer Subtype Identification

Joint Learning of Spectral Clustering and Low-Rank Representation Based on Precise Segmentation Cost for Cancer Subtype Identification

Smart DNA sensors‐based molecular identification for cancer subtyping

AbstractMolecular subtyping of cancer can greatly help to understand the development of disease and predict tumor behavior. Exploring detection methods for precise subtyping is appealing to prognosis and personalized therapy. During the past decades, DNA‐based biosensors have exhibited great potential in cancer diagnosis due to their structural programmability and functional diversity. Despite the encouraging progress that has been made, there remains an issue in improving the accuracy and sensitivity of cancer subtyping due to the complex process of disease, especially in preclinical or clinical applications. To accelerate the development of DNA sensors in the identification of cancer subtypes, in this review, we summarized their advances in molecular subtyping by analyzing the heterogeneity in categories and levels of biomarkers between cancer subtypes. The strategies toward genomic and proteomic heterogeneity in cells or on the cell surface, as well as the cancer excretions including extracellular vesicles (EVs) and microRNA (miRNAs) in serum, are summarized. Current challenges and the opportunities of DNA‐based sensors in this field are also discussed.

Identification of bladder cancer subtypes and predictive model for prognosis, immune features, and immunotherapy based on neutrophil extracellular trap-related genes

Bladder cancer is the most common malignant tumor of urinary system, and its morbidity and mortality are increasing rapidly. Although great advances have been made in medical technology in recent years, there is still a lack of effective prognostic and therapeutic methods for bladder cancer. NETs are reticulated DNA structures decorated with various protein substances released extracellularly by neutrophils stimulated by strong signals. Recently, it has been found that NETs are closely related to the growth, metastasis and drug resistance of many types of cancers. However, up to now, the research on the relationship between NETs and bladder cancer is still not enough. In this study, we aimed to conduct a comprehensive analysis of NRGs in bladder cancer tissues to evaluate the relationship between NRGs and prognosis prediction and sensitivity to therapy in patients with bladder cancer. We scored NRGs in each tissue by using ssGSEA, and selected gene sets that were significantly associated with NRGs scores by using the WCGNA algorithm. Based on the expression profiles of NRGs-related genes, NMF clustering analysis was performed to identify different BLCA molecular subtypes. For the differentially expressed genes between subtypes, we used univariate COX regression, LASSO regression and multivariate COX regression to further construct a hierarchical model of BLCA patients containing 10 genes. This model and the nomogram based on this model can accurately predict the prognosis of BLCA patients in multiple datasets. Besides, BLCA patients classified based on this model differ greatly in their sensitivity to immunotherapy and targeted therapies, which providing a reference for individualized treatment of patients with bladder cancer.

DNA walker-powered SERS logic aptasensors for accurate imaging and identification of tumor cells

Identification of tumor cells is of vital importance for accurate diagnosis and treatment of tumors. Herein, DNA walker-powered SERS logic aptasensors were proposed for accurately identifying tumor cells by high-sensitively and specifically SERS imaging membrane proteins on single live cells. The SERS logic aptasensors consist of gold nanoparticles (AuNPs)-based Logic probes (LPs), Network probes (NPs) and EpCAM probes (EPs). As a proof-of-principle, the cells highly-expressed single EpCAM proteins, single MUC1 proteins, and both or neither the two proteins were tested respectively, and the SERS logic aptasensors can perform AND Boolean logic analysis by EpCAM-triggered DNA walker-powered assembly of AuNP network nanostructures (AuNNs) on cell membrane. By imaging SERS signals of single DTNB, single 4-MBA, simultaneous signals (4-MBA, 2-MBT and DTNB), or none, respectively, the cells with or without the high-expression of EpCAM and/or MUC1 proteins can be accurately and logically identified. Furthermore, the aptasensors were expanded to image the EpCAM and HER2 proteins on cells for accurately identifying the subtypes of breast cancer cells. The SERS logic aptasensors can achieve accurate identification and logic analysis of cancer cell classification and subtypes, and have broad potential applications in accurate diagnosis and treatment of tumors.

An Integrated Method Based on Wasserstein Distance and Graph for Cancer Subtype Discovery.

Due to the complexity of cancer pathogenesis at different omics levels, it is necessary to find a comprehensive method to accurately distinguish and find cancer subtypes for cancer treatment. In this paper, we proposed a new cancer multi-omics subtype identification method, which is based on variational autoencoder measured by Wasserstein distance and graph autoencoder (WVGMO). This method depends on two foremost models. The first model is a variational autoencoder measured by Wasserstein distance (WVAE), which is used to extract potential spatial information of each omic data type. The second model is the graph autoencoder (GAE) with the second-order proximity. It has the capability to retain the topological structure information and feature information of the multi-omics data. And then, the identification of cancer subtypes via k-means clustering. Extensive experiments were conducted on seven different cancers based on four omics data from TCGA. The results show that WVGMO provides equivalent or even better results than the most of advanced synthesis methods.

Automated exploitation of deep learning for cancer patient stratification across multiple types.

Recent frameworks based on deep learning have been developed to identify cancer subtypes from high-throughput gene expression profiles. Unfortunately, the performance of deep learning is highly dependent on its neural network architectures which are often hand-crafted with expertise in deep neural networks, meanwhile, the optimization and adjustment of the network are usually costly and time consuming. To address such limitations, we proposed a fully automated deep neural architecture search model for diagnosing consensus molecular subtypes from gene expression data (DNAS). The proposed model uses ant colony algorithm, one of the heuristic swarm intelligence algorithms, to search and optimize neural network architecture, and it can automatically find the optimal deep learning model architecture for cancer diagnosis in its search space. We validated DNAS on eight colorectal cancer datasets, achieving the average accuracy of 95.48%, the average specificity of 98.07%, and the average sensitivity of 96.24%, respectively. Without the loss of generality, we investigated the general applicability of DNAS further on other cancer types from different platforms including lung cancer and breast cancer, and DNAS achieved an area under the curve of 95% and 96%, respectively. In addition, we conducted gene ontology enrichment and pathological analysis to reveal interesting insights into cancer subtype identification and characterization across multiple cancer types. The source code and data can be downloaded from https://github.com/userd113/DNAS-main. And the web server of DNAS is publicly accessible at 119.45.145.120:5001.

GCNA-Cluster: A Gene Co-Expression Network Alignment to Cluster Cancer Patients Algorithm for Identifying Subtypes of Pancreatic Ductal Adenocarcinoma.

Cancer heterogeneity makes it necessary to use different treatment strategies for patients with the same pathological features. Accurate identification of cancer subtypes is a crucial step in this approach. The current studies of pancreatic ductal adenocarcinoma (PDAC) subtypes mainly focus on single genes and ignore the synergistic effects of genes. Here we proposed a network alignment algorithm GCNA-cluster to cluster patients based on gene co-expression networks. We constructed weighted gene co-expression networks for patients and aligned the networks of two patients to estimate the similarity of patients and their cancer subtypes. A scoring function is defined to measure the network alignment result and the score can indicate the similarity between patients. Then, the patients are clustered based on their similarities. We validated the accuracy of the algorithm on the GEO-PDAC dataset with real labels, and the experimental results show that the GCNA-cluster algorithm has better results than classical cancer subtyping algorithms. In addition, the GCNA-cluster algorithm applied to the TCGA-PDAC dataset identified two subtypes based on the Silhouette Coefficient. Biomarkers identified for the PDAC subtypes hint to cell growth, cell cycle or apoptosis as targets for new therapeutic strategies.