Related Experiment Video
Updated: Jul 7, 2026
07:41
Performing Data Mining And Integrative Analysis Of Biomarker in Breast Cancer Using Multiple Publicly Accessible Databases
Published on: May 17, 2019
Predicting features of breast cancer with gene expression patterns
Xuesong Lu1, Xin Lu, Zhigang C Wang
1Bioinformatics Division, TNLIST and Department of Automation, Tsinghua University, Beijing 100084, China. lxs97@mails.tsinghua.edu.cn
Breast Cancer Research and Treatment
|February 26, 2008
Summary
Global gene expression accurately predicts estrogen receptor (ER), HER2 status, and tumor grade in breast cancer. However, it is less accurate for predicting tumor size, lymph node metastasis, or lymphatic-vascular invasion.
Area of Science:
- Genomics
- Oncology
- Biostatistics
Background:
- Gene expression data from microarrays contain significant biological insights.
- Primary breast cancers exhibit diverse biologic, histologic, and anatomic features.
Purpose of the Study:
- To investigate if global gene expression in primary breast cancers can predict key patient disease features.
- To assess the accuracy of gene expression data in predicting biomarkers, histologic characteristics, and stage parameters.
Main Methods:
- Analysis of microarray data from 129 primary breast cancer patients.
- Statistical prediction models and cross-validation error rates were employed.
- Multidimensional scaling (MDS) was used for data visualization.
Main Results:
- Gene expression accurately predicted estrogen receptor (ER) and HER2 status.
- Tumor grade was dichotomized into high-grade and low-grade clusters, with no distinct intermediate group.
- Prediction of tumor size, lymph node metastasis, and lymphatic-vascular invasion (LVI) using gene expression was inaccurate.
Conclusions:
- Global gene expression supports a binary classification of ER, HER2, and grade in breast tumors.
- Metastasis models may depend on inherent biologic differences between breast cancer subtypes rather than solely on gene expression predictions for stage parameters.