Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers
Breast cancer is the most common female malignancy. It has high mortality, primarily due to metastasis and recurrence. Patients with invasive and noninvasive breast cancer require different treatments, so there is an urgent need for predictive tools to guide clinical decision making and avoid overtr...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-01-01
|
Series: | Cells |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-4409/9/2/326 |
_version_ | 1797719182381416448 |
---|---|
author | Chunyu Wang Ning Zhao Linlin Yuan Xiaoyan Liu |
author_facet | Chunyu Wang Ning Zhao Linlin Yuan Xiaoyan Liu |
author_sort | Chunyu Wang |
collection | DOAJ |
description | Breast cancer is the most common female malignancy. It has high mortality, primarily due to metastasis and recurrence. Patients with invasive and noninvasive breast cancer require different treatments, so there is an urgent need for predictive tools to guide clinical decision making and avoid overtreatment of noninvasive breast cancer and undertreatment of invasive cases. Here, we divided the sample set based on the genome-wide methylation distance to make full use of metastatic cancer data. Specifically, we implemented two differential methylation analysis methods to identify specific CpG sites. After effective dimensionality reduction, we constructed a methylation-based classifier using the Random Forest algorithm to categorize the primary breast cancer. We took advantage of breast cancer (BRCA) HM450 DNA methylation data and accompanying clinical data from The Cancer Genome Atlas (TCGA) database to validate the performance of the classifier. Overall, this study demonstrates DNA methylation as a potential biomarker to predict breast tumor invasiveness and as a possible parameter that could be included in the studies aiming to predict breast cancer aggressiveness. However, more comparative studies are needed to assess its usability in the clinic. Towards this, we developed a website based on these algorithms to facilitate its use in studies and predictions of breast cancer invasiveness. |
first_indexed | 2024-03-12T09:02:06Z |
format | Article |
id | doaj.art-ca0cabd93845437fa1df4ccb1440357e |
institution | Directory Open Access Journal |
issn | 2073-4409 |
language | English |
last_indexed | 2024-03-12T09:02:06Z |
publishDate | 2020-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Cells |
spelling | doaj.art-ca0cabd93845437fa1df4ccb1440357e2023-09-02T15:38:27ZengMDPI AGCells2073-44092020-01-019232610.3390/cells9020326cells9020326Computational Detection of Breast Cancer Invasiveness with DNA Methylation BiomarkersChunyu Wang0Ning Zhao1Linlin Yuan2Xiaoyan Liu3School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150080, ChinaSchool of Life Science and Technology, Harbin Institute of Technology, Harbin 150080, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin 300350, ChinaSchool of Computer Science and Technology, Harbin Institute of Technology, Harbin 150080, ChinaBreast cancer is the most common female malignancy. It has high mortality, primarily due to metastasis and recurrence. Patients with invasive and noninvasive breast cancer require different treatments, so there is an urgent need for predictive tools to guide clinical decision making and avoid overtreatment of noninvasive breast cancer and undertreatment of invasive cases. Here, we divided the sample set based on the genome-wide methylation distance to make full use of metastatic cancer data. Specifically, we implemented two differential methylation analysis methods to identify specific CpG sites. After effective dimensionality reduction, we constructed a methylation-based classifier using the Random Forest algorithm to categorize the primary breast cancer. We took advantage of breast cancer (BRCA) HM450 DNA methylation data and accompanying clinical data from The Cancer Genome Atlas (TCGA) database to validate the performance of the classifier. Overall, this study demonstrates DNA methylation as a potential biomarker to predict breast tumor invasiveness and as a possible parameter that could be included in the studies aiming to predict breast cancer aggressiveness. However, more comparative studies are needed to assess its usability in the clinic. Towards this, we developed a website based on these algorithms to facilitate its use in studies and predictions of breast cancer invasiveness.https://www.mdpi.com/2073-4409/9/2/326breast cancermetastasisinvasivenessdna methylation |
spellingShingle | Chunyu Wang Ning Zhao Linlin Yuan Xiaoyan Liu Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers Cells breast cancer metastasis invasiveness dna methylation |
title | Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers |
title_full | Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers |
title_fullStr | Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers |
title_full_unstemmed | Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers |
title_short | Computational Detection of Breast Cancer Invasiveness with DNA Methylation Biomarkers |
title_sort | computational detection of breast cancer invasiveness with dna methylation biomarkers |
topic | breast cancer metastasis invasiveness dna methylation |
url | https://www.mdpi.com/2073-4409/9/2/326 |
work_keys_str_mv | AT chunyuwang computationaldetectionofbreastcancerinvasivenesswithdnamethylationbiomarkers AT ningzhao computationaldetectionofbreastcancerinvasivenesswithdnamethylationbiomarkers AT linlinyuan computationaldetectionofbreastcancerinvasivenesswithdnamethylationbiomarkers AT xiaoyanliu computationaldetectionofbreastcancerinvasivenesswithdnamethylationbiomarkers |