Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2015 Jun 5;5:11085. doi: 10.1038/srep11085

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

Copyright © 2015, Macmillan Publishers Limited

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

PMC Copyright notice

① Construct the training and test set by sampling patients. The training set and test set contain 80% and 20% of the initial patients for each subtype, respectively. ② Binary classification: LA, LB and TN are classified as one category, called nHER2 (take HER2 as an example). ③ Randomly sample patients, 80% HER2 and 80% nHER2, from the training set, called the outer set. ④ Grow a tree on the particular set obtained from step 2. ⑤ After repeating steps 2-4 1,000 times, we obtained 1,000 trees. ⑥ Count the occurrence frequencies of all features in the above 1,000 trees. ⑦ Randomly assign a label to each patient. Then, use the shuffled data to conduct steps 1-5. Compute the relevance intensity, (FV) , for each feature. Repeat the permutation 10,000 times to obtain an empirical threshold for the specified significance level of β. ⑧ After repeating steps 2–6 1,000 times, we obtained the stable high occurrence frequency of features. ⑨ Construct the model based on the selected features.

Inline graphic — ① Construct the training and test set by sampling patients. The training set and test set contain 80% and 20% of the initial patients for each subtype, respectively. ② Binary classification: LA, LB and TN are classified as one category, called nHER2 (take HER2 as an example). ③ Randomly sample patients, 80% HER2 and 80% nHER2, from the training set, called the outer set. ④ Grow a tree on the particular set obtained from step 2. ⑤ After repeating steps 2-4 1,000 times, we obtained 1,000 trees. ⑥ Count the occurrence frequencies of all features in the above 1,000 trees. ⑦ Randomly assign a label to each patient. Then, use the shuffled data to conduct steps 1-5. Compute the relevance intensity, (FV) , for each feature. Repeat the permutation 10,000 times to obtain an empirical threshold for the specified significance level of β. ⑧ After repeating steps 2–6 1,000 times, we obtained the stable high occurrence frequency of features. ⑨ Construct the model based on the selected features.