Make a list of data batches.
Meta analysis of survival data
Shuffle samples.
Build a design matrix
Z-score2 normalization prior to AUC plot.
Merge survival times and censoring status.
Meta analysis of survival data.
calPerformance.single.indep
Performance assessment on single data sets using independent validation
Exclude samples
Determine the name of a file.
Split data for meta analysis.
Independent validation of the performance of the gene signatures derived from single data sets.
Combine data for meta analysis.
Likelihood function.
ComBat-adjusted microarray gene expression data
Calculate empirical hyper-prior values
Determine the batch ID of data sets.
Performance evaluation by subsetting data sets in 100 iterations
Cross validation with ComBat adjustment
Fit the L/S model in the presence of missing data values
Initiate ComBat adjustment
Pair-wise combination of single data sets prior to the application of ComBat and independent validation.
Iterative solution for Empirical Bayesian method.
Prediction of survival time by independent validation.
Likelihood function.
Generate survival data.
Initiation to build the design matrix
Z-score normalization.
Apply Z-score2 normalization.
Assess the performance obtained from the merged data set by independent validation
Geometric Mean
Calculate empirical hyper-prior values of Bayesian model
Combination of data sets prior to the application of ComBat.
Integration function to find nonparametric adjustments
Plot time-dependent ROC curves from 0 to 120 months.
Reformat gene expression data for ComBat.
Split a data set for cross-validation
Performance assessment of gene signatures by cross-validation.
Matrix Z-score normalization.
Cross validation with or without Z-score normalization
Merge data sets by Z-score2 normalization and assess the performance by independent validation.
Performance assessment of merged data sets by independent validation
Estimated multiplicative batch effect
Cox coefficient calculation.
excl.missing.single.indep
Exclude missing samples prior to independent validation
Merge the data sets by ComBat or Z-score1 normalization and apply independent validation.
Plot ROC curves related to different time points.
Write batch samples for ComBat.
Apply a feature selection
Apply the inverse normal method.
Performance evaluation derived from a subset of a data set
Confidence interval of a Geometric mean
Filter absent calls
Estimated additive batch effect
Determine the indices of the training or testing set.
Performance evaluation by merging two simulated independent data sets
Exclude missing samples
Plot ROC curves of the testing set normalized by a joint analysis method.
Feature selection for meta analysis
calPerformance.merge.indep
Assess performance derived from the merged data set by independent validation
Survival Prediction by Joint Analysis of Microarray Gene Expression Data
Determine the indices of the training and testing sets.
Trim the data.
Merge data set by ComBat within cross-validation.
Apply Z-score1 normalization.
Simulate survival data.
Start plotting
main.process