XGBoost-based and tumor-immune characterized gene signature for the prediction of metastatic status in breast cancer

Li, Qingqing; Yang, Hui; Wang, Peipei; Liu, Xiaocen; Lv, Kun; Ye, Mingquan

doi:10.1186/s12967-022-03369-9

Cited by 42 publications

(33 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Logistic regression is the traditional method of predicting disease, and a Nomogram is a graph that lists the level of risk for each metric. In the past decade, the nomogram has been accepted as a reliable method for predicting tumor prognosis [9] . It has been applied to the prognosis prediction of many cancers, including gastric cancer, breast cancer, testicular cancer, etc [10][11][12][13] .…”

Section: Introductionmentioning

confidence: 99%

Construction of XGBoost Model for Early Lung Cancer Prediction Based on Metabolic Indexes

Guan

Teng

et al. 2022

Preprint

View full text Add to dashboard Cite

Background In order to test the predictive effect of the XGBoost model in the machine learning algorithm for early lung cancer diagnosis and the importance of amino acid and carnitine indicators in affecting the occurrence of lung cancer, this study used data from 848 patients with lung cancer and benign lung nodules. Results The study selected 49 serum amino acid and carnitine indexes, as well as age and gender demographic indexes of subjects from a hospital in Dalian City, Liaoning Province. After screening with stepwise regression algorithm, 16 indicators were included.Draw the ROC curve of 5 models, comparing the accuracy, precision, recall, and F1 score of the lung cancer prediction model with the nomogram and the machine learning algorithm, the optimal model (XGBoost) was obtained, and the feature importance ranking was obtained using the XGBoost model.In the method of screening indicators, the index modeling accuracy after screening using the stepwise regression algorithm was the highest, which was 75.29%, and the AUC value was 0.81. In the model performance comparison, the accuracy of the nomogram is 68.24%, and the AUC value is 0.74. Among the machine learning models, the XGBoost algorithm has the best performance, with an accuracy rate of 75.29% and an ROC value of 0.81. In the model feature importance score results, the top three important features were ornithine, valine and palmitoylcarnitine. Conclusions The combined use of nomogram and machine learning model(XGBoost) not only improves the accuracy of the lung cancer prediction model, but also makes the understanding of the results more intuitive and convenient for doctors and patients.

show abstract

Section: Introductionmentioning

confidence: 99%

Construction of XGBoost Model for Early Lung Cancer Prediction Based on Metabolic Indexes

Guan

Teng

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Using weighted gene co-expression network analysis, Wang et al (2020) identified six hub genes associated with IA rupture, which tended to enrich in one pathway. In addition, the gene with the highest ranking for classification importance was not necessarily the one with the greatest fold change and vice versa ( Li et al, 2022 ). The feature selection algorithm had the benefit over conventional statistical methods in this regard, favoring the selection of biomarkers that could be used to distinguish between IAs that had ruptured and those that had not.…”

Section: Discussionmentioning

confidence: 99%

A two-stage hybrid gene selection algorithm combined with machine learning models to predict the rupture status in intracranial aneurysms

Wang

Yuan

et al. 2022

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

An IA is an abnormal swelling of cerebral vessels, and a subset of these IAs can rupture causing aneurysmal subarachnoid hemorrhage (aSAH), often resulting in death or severe disability. Few studies have used an appropriate method of feature selection combined with machine learning by analyzing transcriptomic sequencing data to identify new molecular biomarkers. Following gene ontology (GO) and enrichment analysis, we found that the distinct status of IAs could lead to differential innate immune responses using all 913 differentially expressed genes, and considering that there are numerous irrelevant and redundant genes, we propose a mixed filter- and wrapper-based feature selection. First, we used the Fast Correlation-Based Filter (FCBF) algorithm to filter a large number of irrelevant and redundant genes in the raw dataset, and then used the wrapper feature selection method based on the he Multi-layer Perceptron (MLP) neural network and the Particle Swarm Optimization (PSO), accuracy (ACC) and mean square error (MSE) were then used as the evaluation criteria. Finally, we constructed a novel 10-gene signature (YIPF1, RAB32, WDR62, ANPEP, LRRCC1, AADAC, GZMK, WBP2NL, PBX1, and TOR1B) by the proposed two-stage hybrid algorithm FCBF-MLP-PSO and used different machine learning models to predict the rupture status in IAs. The highest ACC value increased from 0.817 to 0.919 (12.5% increase), the highest area under ROC curve (AUC) value increased from 0.87 to 0.94 (8.0% increase), and all evaluation metrics improved by approximately 10% after being processed by our proposed gene selection algorithm. Therefore, these 10 informative genes used to predict rupture status of IAs can be used as complements to imaging examinations in the clinic, meanwhile, this selected gene signature also provides new targets and approaches for the treatment of ruptured IAs.

show abstract

“…IL32 is a pro-inflammatory cytokine that is highly expressed in malignant cells and can promote their invasion and metastasis in a variety of cancer types [37][38][39]. TMEM64 is a regulator of Wnt/β-catenin signaling pathway, which was found to be associated with metastasis in prostate and breast cancers [40,41].…”

Section: Discussionmentioning

confidence: 99%

Developing high-resolution metastasis signatures for improved cancer prognosis using single-cell RNA sequencing data：A case study in lung adenocarcinoma

Zhou

Zhang

et al. 2022

Preprint

View full text Add to dashboard Cite

Background Metastasis remains the reason for high cancer mortality and it is a valuable predictive factor in cancer prognosis. Single-cell RNA sequencing (scRNA-seq) can reveal cellular heterogeneity in metastasis microenvironment and capture high-resolution signatures for improved cancer prediction. Methods An integrated analysis framework was designed for metastatic lung adenocarcinoma (LUAD) scRNA-seq profiles and we identified 9 key prognostic genes (KPGs) that were trained and validated in 407 internal and external patient cohorts using Lasso-Cox method and Receiver Operating Characteristic (ROC) curves. To ensure the predictive stability of the KPGs signatures, 10 random samples of data from the TCGA cohort were taken. Correlation analysis revealed the strong association between KPGs signatures and several clinical characteristics such as gender, T-stage, and N-stage. We incorporated these risk clinical variables into a KPGs nomogram model. Results The results based on ROC curves and calibration curves show that the KPGs nomogram model with superior accuracy for overall survival (OS) prediction. We also found that high risk group with high nomogram scores had poorer prognosis accompanied by a higher tumor mutation burden (TMB) and it was associated with the upregulation of cell cycle, DNA replication, ECM receptor interaction, P53 signaling pathway, spliceosome and proteasome pathway. Conclusions Mining single-cell resolution metastatic features from scRNA-seq data to improve cancer prognosis is a viable strategy that would be a useful tool in risk gene discovery and targeted therapy in metastatic cancers.

show abstract

XGBoost-based and tumor-immune characterized gene signature for the prediction of metastatic status in breast cancer

Cited by 42 publications

References 49 publications

Construction of XGBoost Model for Early Lung Cancer Prediction Based on Metabolic Indexes

Construction of XGBoost Model for Early Lung Cancer Prediction Based on Metabolic Indexes

A two-stage hybrid gene selection algorithm combined with machine learning models to predict the rupture status in intracranial aneurysms

Developing high-resolution metastasis signatures for improved cancer prognosis using single-cell RNA sequencing data：A case study in lung adenocarcinoma

Contact Info

Product

Resources

About