<Warped linear mixed models for the genetic analysis of transformed phenotypes>是篇讲对数据正态化的新方法的文章,新方法将就叫做弯曲混合线性模型.地址:http://www.nature.com/ncomms/2014/140919/ncomms5890/full/ncomms5890.html.
文章最后给出了一个python模块,地址https://github.com/pmbio/warpedLMM.
可使用下面命令安装.
sudo pip install warpedlmm
可惜安装好之后,问题不断.
1.模型依赖GPy模型,不断提示UserWarning: warning: caught this exception:'module' object has no attribute '_dotblas',目前没有解决,鉴于只是警告,暂时忽略.
2.warpedlmm.py文件18和19行载入语句错误,
import transformations from warping_functions import TanhWarpingFunction_d
应为:
import util.transformations as transformations from util.warping_functions import TanhWarpingFunction_d
3.模型说明文件中依赖库有'scipy>=0.13', 'numpy>=1.8', 'matplotlib>=1.2', 'nose', 'fastlmm',但是fastlmm模型依赖sklearn,故要手动安装
pip install -U scikit-leran