https://www.ncbi.nlm.nih.gov/pubmed/30742607
Network-guided prediction of aromatase inhibitor response in breast cancer.
Author information
- 1
- Computational Biology Department, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States of America.
- 2
- Women's Cancer Research Center, Department of Pharmacology and Chemical Biology, UPMC Hillman Cancer Center, Magee Womens Research Institute, Pittsburgh, Pennsylvania, United States of America.
- 3
- Machine Learning Department, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States of America.
Abstract
Prediction of response to specific cancer treatments is complicated by significant heterogeneity between tumors in terms of mutational profiles, gene expression, and clinical measures. Here we focus on the response of Estrogen Receptor (ER)+ post-menopausal breast cancer tumors to aromatase inhibitors (AI). We use a network smoothing algorithm to learn novel features that integrate several types of high throughput data and new cell line experiments. These features greatly improve the ability to predict response to AI when compared to prior methods. For a subset of the patients, for which we obtained more detailed clinical information, we can further predict response to a specific AI drug.
- PMID:
- 30742607
- PMCID:
- PMC6386390
- DOI:
- 10.1371/journal.pcbi.1006730
카네기 멜론 대학에서 낸 저널이니 봐야함.
PlosOne에 등록된 것도 그렇고
유방암.
폐경이후의 에스트로젠은, 말초조직에서, aromatase enzyme에 의해서 생성
ER+ 인 유방암에서, aromatase inhibitor 치료시, 이 치료약에 대한 반응은, 사람마다 다 다를 수 밖에.
그래서 aromatase inhibitor 를 누구에게 쓸지 (누구에서 쓸 때 더 효과적인지)를 결정하는 것은, 유방암 치료에 중요한 일.
이것을, tumor omic data (cell line experiments) 를 컴바인해서, high-quality clinical data랑, machine learning 방법으로 예측하는 모델을 만듬.
방법 : Network smoothing 방법. ( network proximity
measures by computing the element-wise minimum of the smoothed scores )
- probabilistic SVM and Random Forest (RF) classifiers
( using the scikit-learn package )
- high scoring PCA features
- leave-one-out cross-validation folds for random forest classifiers
변수들 : somatic mutations,
differentially expressed genes,
and protein targets for a particular drug
.....
......
임상자료들 :
reproductive history of patients at the time of breast cancer diagnosis,
family history including both first and second degree relatives as well as other malignancy history for the patients if applicable
hormone receptor status including H-score or percent staining as well as HER2/neu status
AI drug ... 여기서 AI는 aromatase inhibitor 임.. ;;;;
딥러닝이나 강화학습을 이용한 건 이니고
기존에 성능이 좋은 SVM, Random Forrest 등을 이용하고
좋은 classifier 를 써서
양질의 임상자료랑, 유전학적인 cell line genomic data 를
합쳐서, aromatase inhibitor(AI) 에 반응이 좋을 것 같은
성능 좋은 예측 모델을 만들었다는...
그런 논문인듯.
2019 PLoSone_Network-guided prediction of aromatase inhibitor response in breast cancer.pdf
