Robust Gene-Environment Interaction Analysis
Extract coefficients from a "bic.BLMCP" object
Extract coefficients from a "bic.PTReg" object
Extract coefficients from a "BLMCP" object
Extract coefficients from a "PTReg" object
Extract coefficients from a "RobSBoosting" object
Robust gene-environment interaction analysis approach via sparse boost...
Plot coefficients from a "bic.BLMCP" object
Plot coefficients from a "bic.PTReg" object
Plot coefficients from a "BLMCP" object
Plot coefficients from a "Miss.boosting" object
Plot coefficients from a "PTReg" object
Plot coefficients from a "RobSBoosting" object
Make predictions from a "bic.BLMCP" object.
Make predictions from a "bic.PTReg" object
Make predictions from a "BLMCP" object
Make predictions from a "Miss.boosting" object
Make predictions from a "PTReg" object
Make predictions from a "RobSBoosting" object
Robust gene-environment interaction analysis using penalized trimmed r...
Robust identification of gene-environment interactions using a quantil...
Simulated data for generating response
P-values of the "QPCorr.matrix" obtained using a permutation approach
Robust semiparametric gene-environment interaction analysis using spar...
The covariance matrix with an autoregressive (AR) structure among vari...
Accommodating missingness in environmental measurements in gene-enviro...
BIC for BLMCP
BIC for PTReg
Accommodating missingness in environmental measurements in gene-enviro...
Description: For the risk, progression, and response to treatment of many complex diseases, it has been increasingly recognized that gene-environment interactions play important roles beyond the main genetic and environmental effects. In practical interaction analyses, outliers in response variables and covariates are not uncommon. In addition, missingness in environmental factors is routinely encountered in epidemiological studies. The developed package consists of five robust approaches to address the outliers problems, among which two approaches can also accommodate missingness in environmental factors. Both continuous and right censored responses are considered. The proposed approaches are based on penalization and sparse boosting techniques for identifying important interactions, which are realized using efficient algorithms. Beyond the gene-environment analysis, the developed package can also be adopted to conduct analysis on interactions between other types of low-dimensional and high-dimensional data. (Mengyun Wu et al (2017), <doi:10.1080/00949655.2018.1523411>; Mengyun Wu et al (2017), <doi:10.1002/gepi.22055>; Yaqing Xu et al (2018), <doi:10.1080/00949655.2018.1523411>; Yaqing Xu et al (2019), <doi:10.1016/j.ygeno.2018.07.006>; Mengyun Wu et al (2021), <doi:10.1093/bioinformatics/btab318>).