A personalized committee classification approach to improving prediction of breast cancer metastasis
نویسندگان
چکیده
MOTIVATION Metastasis prediction is a well-known problem in breast cancer research. As breast cancer is a complex and heterogeneous disease with many molecular subtypes, predictive models trained for one cohort often perform poorly on other cohorts, and a combined model may be suboptimal for individual patients. Furthermore, attempting to develop subtype-specific models is hindered by the ambiguity and stereotypical definitions of subtypes. RESULTS Here, we propose a personalized approach by relaxing the definition of breast cancer subtypes. We assume that each patient belongs to a distinct subtype, defined implicitly by a set of patients with similar molecular characteristics, and construct a different predictive model for each patient, using as training data, only the patients defining the subtype. To increase robustness, we also develop a committee-based prediction method by pooling together multiple personalized models. Using both intra- and inter-dataset validations, we show that our approach can significantly improve the prediction accuracy of breast cancer metastasis compared with several popular approaches, especially on those hard-to-learn cases. Furthermore, we find that breast cancer patients belonging to different canonical subtypes tend to have different predictive models and gene signatures, suggesting that metastasis in different canonical subtypes are likely governed by different molecular mechanisms. AVAILABILITY AND IMPLEMENTATION Source code implemented in MATLAB and Java available at www.cs.utsa.edu/∼jruan/PCC/.
منابع مشابه
A Personalized Committee Classification Approach to Improving Prediction of Breast Cancer Metastasis – Supplementary Materials
متن کامل
Prediction of Breast Cancer Metastasis Using Fuzzy Models based on Data from Iranian Breast Cancer Patients
Introduction: The metastasis of breast cancer, the spread of cancer to different body parts, is considered as one of the most important factors responsible for the majority of deaths caused by breast cancer in women. Diagnosing the breast cancer metastasis at the earliest stages helps to choose the best treatment and improve the quality of life for patients. Method: In the present fundamental r...
متن کاملPrediction of Breast Cancer Metastasis Using Fuzzy Models based on Data from Iranian Breast Cancer Patients
Introduction: The metastasis of breast cancer, the spread of cancer to different body parts, is considered as one of the most important factors responsible for the majority of deaths caused by breast cancer in women. Diagnosing the breast cancer metastasis at the earliest stages helps to choose the best treatment and improve the quality of life for patients. Method: In the present fundamental r...
متن کاملBioinformatics-Based Prediction of FUT8 as a Therapeutic Target in Estrogen Receptor-Positive Breast Cancer
Abstract Introduction: Estrogen receptor-positive (ER-positive) breast cancer is a subgroup of breast tumors that is more likely to respond to hormone therapy. ER-positive and ER- negative breast cancers tend to show different patterns of metastasis because of different signaling cascade and genes that are activated by estrogen response. Genetic factors can contribute to high rates of metastas...
متن کاملLymph Node Ratio is More Predictive than Traditional Lymph Node Stratification for Invasive Breast Cancer
Over the past three decades, the breast cancer (BC) incidence has been steadily increasing and becoming the most common malignancy in large cities, like Shanghai (Fan et al., 2009) in China. Accurate evaluation for each patient is fundamental for BC personalized care. TNM staging system is the essential classification for BC treatment decision and prognosis prediction over the past 60 years, wh...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Bioinformatics
دوره 30 13 شماره
صفحات -
تاریخ انتشار 2014