+
+Show the index of selected variables by controlling Bayesian false discovery rate (FDR) at the level $\alpha = 0.05$
+
+```r
+which( VS(fit, method = "FDR", threshold = 0.05) )
+```
+```{ .text .no-copy }
+#[1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 194
+```
+
+
+### Plot time-dependent Brier scores
+
+The function `BayesSurvive::plotBrier()` can show the time-dependent Brier scores based on posterior mean of coefficients or Bayesian model averaging.
+
+```r
+plotBrier(fit, survObj.new = dataset)
+```
+
+
+
+We can also use the function `BayesSurvive::predict()` to obtain the Brier score at time 8.5, the integrated Brier score (IBS) from time 0 to 8.5 and the index of prediction accuracy (IPA).
+
+```r
+predict(fit, survObj.new = dataset, times = 8.5)
+```
+```{ .text .no-copy }
+## Brier(t=8.5) IBS(t:0~8.5) IPA(t=8.5)
+## Null.model 0.2290318 0.08185316 0.0000000
+## Bayesian.Cox 0.1013692 0.02823275 0.5574011
+```
+
+### Predict survival probabilities and cumulative hazards
+
+The function `BayesSurvive::predict()` can estimate the survival probabilities and cumulative hazards.
+
+```r
+predict(fit, survObj.new = dataset, type = c("cumhazard", "survival"))
+```
+```{ .text .no-copy }
+# observation times cumhazard survival
+## Bayesian Cox Models with graph-structure +priors
+ + + +This is a R/Rcpp package BayesSurvive for Bayesian +survival models with graph-structured selection priors for sparse +identification of high-dimensional features predictive of survival (Madjar +et al., 2021) and its extensions with the use of a fixed graph via a +Markov Random Field (MRF) prior for capturing known structure of +high-dimensional features, e.g. disease-specific pathways from the Kyoto +Encyclopedia of Genes and Genomes (KEGG) database.
+
+
+Installation
+Install the latest released version from CRAN
+ +Install the latest development version from GitHub
+ +
+
+Examples
+
+
+Simulate data
+ +
+
+
+Run a Bayesian Cox model
+## Initial value: null model without covariates
+initial = list("gamma.ini" = rep(0, ncol(dataset$X)))
+# Prior parameters
+hyperparPooled = list(
+ "c0" = 2, # prior of baseline hazard
+ "tau" = 0.0375, # sd (spike) for coefficient prior
+ "cb" = 20, # sd (slab) for coefficient prior
+ "pi.ga" = 0.02, # prior variable selection probability for standard Cox models
+ "a" = -4, # hyperparameter in MRF prior
+ "b" = 0.1, # hyperparameter in MRF prior
+ "G" = simData$G # hyperparameter in MRF prior
+)
+
+## run Bayesian Cox with graph-structured priors
+set.seed(123)
+fit <- BayesSurvive(survObj = dataset, model.type = "Pooled", MRF.G = TRUE,
+ hyperpar = hyperparPooled, initial = initial,
+ nIter = 200, burnin = 100)
+
+## show posterior mean of coefficients and 95% credible intervals
+library("GGally")
+plot(fit) +
+ coord_flip() +
+ theme(axis.text.x = element_text(angle = 90, size = 7))
Show the index of selected variables by controlling Bayesian false +discovery rate (FDR) at the level \(\alpha = +0.05\)
+ +#[1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 194
+
+Plot time-dependent Brier scores
+The function BayesSurvive::plotBrier() can show the
+time-dependent Brier scores based on posterior mean of coefficients or
+Bayesian model averaging.
We can also use the function BayesSurvive::predict() to
+obtain the Brier score at time 8.5, the integrated Brier score (IBS)
+from time 0 to 8.5 and the index of prediction accuracy (IPA).
## Brier(t=8.5) IBS(t:0~8.5) IPA(t=8.5)
+## Null.model 0.2290318 0.08185316 0.0000000
+## Bayesian.Cox 0.1013692 0.02823275 0.5574011
+
+Predict survival probabilities and cumulative hazards
+The function BayesSurvive::predict() can estimate the
+survival probabilities and cumulative hazards.
# observation times cumhazard survival
+## <int> <num> <num> <num>
+## 1: 1 3.3 7.41e-05 1.00e+00
+## 2: 2 3.3 2.51e-01 7.78e-01
+## 3: 3 3.3 9.97e-07 1.00e+00
+## 4: 4 3.3 1.84e-03 9.98e-01
+## 5: 5 3.3 3.15e-04 1.00e+00
+## ---
+## 9996: 96 9.5 7.15e+00 7.88e-04
+## 9997: 97 9.5 3.92e+02 7.59e-171
+## 9998: 98 9.5 2.81e+00 6.02e-02
+## 9999: 99 9.5 3.12e+00 4.42e-02
+## 10000: 100 9.5 1.97e+01 2.79e-09
+
+Run a ‘Pooled’ Bayesian Cox model with graphical learning
+ +
+
+
+
+Run a Bayesian Cox model with subgroups using fixed graph
+# specify a fixed joint graph between two subgroups
+hyperparPooled$G <- Matrix::bdiag(simData$G, simData$G)
+dataset2 <- simData[1:2]
+dataset2 <- lapply(dataset2, setNames, c("X", "t", "di", "X.unsc", "trueB"))
+fit3 <- BayesSurvive(survObj = dataset2,
+ hyperpar = hyperparPooled, initial = initial,
+ model.type="CoxBVSSL", MRF.G = TRUE,
+ nIter = 10, burnin = 5)
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/man/VS.Rd b/man/VS.Rd
index b1fee5d..1c8aae2 100644
--- a/man/VS.Rd
+++ b/man/VS.Rd
@@ -7,7 +7,8 @@
VS(x, method = "FDR", threshold = NA, subgroup = 1)
}
\arguments{
-\item{x}{fitted object obtained with \code{BayesSurvive}}
+\item{x}{fitted object obtained with \code{BayesSurvive}, or a matrix/array,
+or a list consisting of matrices and arrays}
\item{method}{variable selection method to choose from
\code{c("CI", "SNC", "MPM", "FDR")}. Default is "FDR"}
diff --git a/man/predict.BayesSurvive.Rd b/man/predict.BayesSurvive.Rd
index f5a101b..793b88a 100644
--- a/man/predict.BayesSurvive.Rd
+++ b/man/predict.BayesSurvive.Rd
@@ -43,10 +43,10 @@ Cox model}
}
\value{
A list object with 5 components if \code{type="brier"} including
-"model", "times", "Brier", "IBS" and "IPA" (Index of Prediction Accuracy),
+"model", "times", "Brier", "IBS" and "IPA" (Index of Prediction Accuracy),
otherwise a list of 7 components with the first component as
-the specified argument \code{type} and "se", "band", "type", "diag",
-"baseline" and "times", see function \code{riskRegression::predictCox} for
+the specified argument \code{type} and "se", "band", "type", "diag",
+"baseline" and "times", see function \code{riskRegression::predictCox} for
details
}
\description{
diff --git a/src/RcppExports.cpp b/src/RcppExports.cpp
index a93b284..cc51396 100644
--- a/src/RcppExports.cpp
+++ b/src/RcppExports.cpp
@@ -41,30 +41,6 @@ BEGIN_RCPP
return rcpp_result_gen;
END_RCPP
}
-// matProdVec
-arma::mat matProdVec(const arma::mat x, const arma::vec y);
-RcppExport SEXP _BayesSurvive_matProdVec(SEXP xSEXP, SEXP ySEXP) {
-BEGIN_RCPP
- Rcpp::RObject rcpp_result_gen;
- Rcpp::RNGScope rcpp_rngScope_gen;
- Rcpp::traits::input_parameter< const arma::mat >::type x(xSEXP);
- Rcpp::traits::input_parameter< const arma::vec >::type y(ySEXP);
- rcpp_result_gen = Rcpp::wrap(matProdVec(x, y));
- return rcpp_result_gen;
-END_RCPP
-}
-// sumMatProdVec
-arma::vec sumMatProdVec(const arma::mat x, const arma::vec y);
-RcppExport SEXP _BayesSurvive_sumMatProdVec(SEXP xSEXP, SEXP ySEXP) {
-BEGIN_RCPP
- Rcpp::RObject rcpp_result_gen;
- Rcpp::RNGScope rcpp_rngScope_gen;
- Rcpp::traits::input_parameter< const arma::mat >::type x(xSEXP);
- Rcpp::traits::input_parameter< const arma::vec >::type y(ySEXP);
- rcpp_result_gen = Rcpp::wrap(sumMatProdVec(x, y));
- return rcpp_result_gen;
-END_RCPP
-}
// updateBH_cpp
arma::vec updateBH_cpp(const arma::mat x_, const arma::vec beta_, const unsigned int J_, const arma::mat ind_r_d_, const arma::vec hPriorSh_, const arma::vec d_, const double c0_);
RcppExport SEXP _BayesSurvive_updateBH_cpp(SEXP x_SEXP, SEXP beta_SEXP, SEXP J_SEXP, SEXP ind_r_d_SEXP, SEXP hPriorSh_SEXP, SEXP d_SEXP, SEXP c0_SEXP) {
diff --git a/src/init.c b/src/init.c
index d03c0ac..acdebc3 100644
--- a/src/init.c
+++ b/src/init.c
@@ -3,29 +3,25 @@
#include References
+++Katrin Madjar, Manuela Zucknick, Katja Ickstadt, Jörg Rahnenführer +(2021). Combining heterogeneous subgroups with graph‐structured variable +selection priors for Cox regression. BMC Bioinformatics, +22(1):586. DOI: 10.1186/s12859-021-04483-z.
+