Ready for CRAN

Author: fbertran

vignettes/benchmarking-selectboost-networks.Rmd

@@ -16,6 +16,8 @@ vignette: >
 ---
 
 ```{r setup, include = FALSE}
+LOCAL <- identical(Sys.getenv("LOCAL"), "TRUE")
+knitr::opts_chunk$set(purl = LOCAL)
 knitr::opts_chunk$set(
   collapse = TRUE,
   comment = "#>"
@@ -41,22 +43,22 @@ The following allows to reproduce some the figures of the article Aouadi et al.
 
 ## Retrieve the results
 Some simulation results were stored in the `results_simuls_reverse_engineering_v3` dataset provided with the package. 
-```{r loadresults, cache= FALSE}
+```{r loadresults, cache= FALSE, eval = LOCAL}
 library(SelectBoost)
 data(results_simuls_reverse_engineering_v3)
 colgrey=grey(.05,NULL)
 ```
 
 ## Compute graphs' ranges
-```{r ranges, cache= FALSE}
+```{r ranges, cache= FALSE, eval = LOCAL}
 rangeCPy_S=range(sensitivity_C,sensitivity_PL,sensitivity_PL2,sensitivity_PL2_W,sensitivity_PL2_tW,sensitivity_PSel,sensitivity_PSel_W,sensitivity_PSel.5,sensitivity_PSel.e2,sensitivity_PSel.5.e2,sensitivity_robust,sensitivity_PB,predictive_positive_value_PB_095_075,predictive_positive_value_PB_075_075,sensitivity_PB_W)
 rangeCPy_PPV=range(predictive_positive_value_C,predictive_positive_value_PL,predictive_positive_value_PL2,predictive_positive_value_PL2_W,predictive_positive_value_PL2_tW,predictive_positive_value_PSel,predictive_positive_value_PSel_W,predictive_positive_value_PSel.5,predictive_positive_value_PSel.e2,predictive_positive_value_PSel.5.e2,predictive_positive_value_robust,predictive_positive_value_PB,predictive_positive_value_PB_095_075,predictive_positive_value_PB_075_075,predictive_positive_value_PB_W)
 rangeCPy_F=range(F_score_C,F_score_PL,F_score_PL2,F_score_PL2_W,F_score_PL2_tW,F_score_PSel,F_score_PSel_W,F_score_PSel.5,F_score_PSel.e2,F_score_PSel.5.e2,F_score_PB,F_score_PB_095_075,F_score_PB_075_075,F_score_PB_W)
 rangeCPx=range(test.seq_C,test.seq_PL,test.seq_PL2,test.seq_PL2_W,test.seq_PL2_tW,test.seq_PSel,test.seq_PSel_W,test.seq_PSel.5,test.seq_PSel.e2,test.seq_PSel.5.e2,test.seq_robust,test.seq_PB,test.seq_PB_095_075,test.seq_PB_075_075,test.seq_PB_W)
 ```
 
 ## Sensitivity
-```{r artgraphs1, cache= FALSE, fig.width=6}
+```{r artgraphs1, cache= FALSE, fig.width=6, eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_PL2),t(sensitivity_PL2),type="l",xlab="cutoff",ylab="Sensitivity",main="Patterns_Lasso2",ylim=rangeCPy_S,col=grey(.05,NULL))
 abline(v=nv_PL2,col=grey(.05,NULL),lty=3)
@@ -73,7 +75,7 @@ abline(v=nv_PB_W,col=grey(.05,NULL),lty=3)
 ```
 
 Run the code to plot the graph.
-```{r artgraphs2, cache= FALSE, fig.width=6, fig.keep='none'}
+```{r artgraphs2, cache= FALSE, fig.width=6, fig.keep='none', eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_robust),t(sensitivity_robust),type="l",xlab="cutoff",ylab="Sensitivity",main="Patterns_Robust",ylim=rangeCPy_S,col=grey(.05,NULL))
 abline(v=nv_robust,col=grey(.05,NULL),lty=3)
@@ -90,7 +92,7 @@ abline(v=nv_PSel,col=grey(.05,NULL),lty=3)
 ```
 
 ## Predictive positive value
-```{r artgraphs3, cache= FALSE, fig.width=6}
+```{r artgraphs3, cache= FALSE, fig.width=6, eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_PL2),t(predictive_positive_value_PL2),type="l",xlab="cutoff",ylab="Predictive Positive Value",main="Patterns_Lasso2",ylim=rangeCPy_PPV,col=grey(.05,NULL))
 abline(v=nv_PL2,col=grey(.05,NULL),lty=3)
@@ -107,7 +109,7 @@ abline(v=nv_PB_W,col=grey(.05,NULL),lty=3)
 ```
 
 Run the code to plot the graph.
-```{r artgraphs4, cache= FALSE, fig.width=6, fig.keep='none'}
+```{r artgraphs4, cache= FALSE, fig.width=6, fig.keep='none', eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_robust),t(predictive_positive_value_robust),type="l",xlab="cutoff",ylab="Predictive Positive Value",main="Patterns_Robust",ylim=rangeCPy_PPV,col=grey(.05,NULL))
 abline(v=nv_robust,col=grey(.05,NULL),lty=3)
@@ -124,7 +126,7 @@ abline(v=nv_PSel,col=grey(.05,NULL),lty=3)
 ```
 
 ## F-score
-```{r artgraphs5, cache= FALSE, fig.width=6}
+```{r artgraphs5, cache= FALSE, fig.width=6, eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_PL2),t(F_score_PL2),type="l",xlab="cutoff",ylab="Fscore",main="Patterns_Lasso2",ylim=rangeCPy_F,col=grey(.05,NULL))
 abline(v=nv_PL2,col=grey(.05,NULL),lty=3)
@@ -141,7 +143,7 @@ abline(v=nv_PB_W,col=grey(.05,NULL),lty=3)
 ```
 
 Run the code to plot the graph.
-```{r artgraphs6, cache= FALSE, fig.width=6, fig.keep='none'}
+```{r artgraphs6, cache= FALSE, fig.width=6, fig.keep='none', eval = LOCAL}
 layout(matrix(1:6,nrow=2))
 matplot(t(test.seq_robust),t(F_score_robust),type="l",xlab="cutoff",ylab="Fscore",main="Patterns_Robust",ylim=rangeCPy_F,col=grey(.05,NULL))
 abline(v=nv_robust,col=grey(.05,NULL),lty=3)

vignettes/confidence-indices-Cascade-networks.Rmd

@@ -16,6 +16,8 @@ vignette: >
 ---
 
 ```{r setup, include = FALSE}
+LOCAL <- identical(Sys.getenv("LOCAL"), "TRUE")
+knitr::opts_chunk$set(purl = LOCAL)
 knitr::opts_chunk$set(
   collapse = TRUE,
   comment = "#>"
@@ -42,12 +44,12 @@ devtools::install_github("fbertran/SelectBoost", ref = "doMC")
 Code to reproduce the datasets saved with the package and some the figures of the article Aouadi et al. (2018), <arXiv:1810.01670>
 
 ## Data simulation
-```{r Cascade, cache= FALSE}
+```{r Cascade, cache= FALSE, eval = LOCAL}
 library(Cascade)
 ```
 
 We define the F array for the simulations.
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 T<-4
 F<-array(0,c(T-1,T-1,T*(T-1)/2))
 
@@ -61,7 +63,7 @@ F[,,5]<-F[,,2]
 ```
 
 We set the seed to make the results reproducible
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 set.seed(1)
 Net<-Cascade::network_random(
   nb=100,
@@ -77,7 +79,7 @@ Net@F<-F
 ```
 
 We simulate gene expression according to the network Net
-```{r message=FALSE, cache=TRUE}
+```{r message=FALSE, cache=TRUE, eval = LOCAL}
 M <- Cascade::gene_expr_simulation(
   network=Net,
   time_label=rep(1:4,each=25),
@@ -87,100 +89,100 @@ M <- Cascade::gene_expr_simulation(
 
 ## Network inference
 We infer the new network.
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 Net_inf_C <- Cascade::inference(M,cv.subjects=TRUE)
 ```
 
 Heatmap of the coefficients of the Omega matrix of the network. Run the code to get the graph.
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 stats::heatmap(Net_inf_C@network, Rowv = NA, Colv = NA, scale="none", revC=TRUE)
 ```
 
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 Fab_inf_C <- Net_inf_C@F
 ```
 
 ## Compute the confidence indices for the inference
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 library(SelectBoost)
 set.seed(1)
 ```
 
 By default the crossvalidation is made subjectwise according to a leave one out scheme and the resampling analysis is made at the .95 `c0` level. To pass CRAN tests, `use.parallel = FALSE` is required. Set `use.parallel = TRUE` and select the number of cores using `ncores = 4`.
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 net_pct_selected <- selectboost(M, Fab_inf_C, use.parallel = FALSE)
 ```
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 net_pct_selected_.5 <- selectboost(M, Fab_inf_C, c0value = .5, use.parallel = FALSE)
 ```
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 net_pct_selected_thr <- selectboost(M, Fab_inf_C, group = group_func_1, use.parallel = FALSE)
 ```
 
 Use `cv.subject=FALSE` to use default crossvalidation
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 net_pct_selected_cv <- selectboost(M, Fab_inf_C, cv.subject=FALSE, use.parallel = FALSE)
 ```
 
 ## Analysis of the confidence indices
 Use plot to display the result of the confidence analysis.
-```{r, cache= TRUE}
+```{r, cache= TRUE, eval = LOCAL}
 plot(net_pct_selected)
 ```
 
 Run the code to plot the other results.
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(net_pct_selected_.5)
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(net_pct_selected_thr)
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(net_pct_selected_cv)
 ```
 
 Run the code to plot the remaning graphs.
 
 Distribution of non-zero (absolute value > 1e-5) coefficients
-```{r, cache= FALSE, fig.keep="none"}
+```{r, cache= FALSE, fig.keep="none", eval = LOCAL}
 hist(Net_inf_C@network[abs(Net_inf_C@network)>1e-5])
 ```
 
 Plot of confidence at .95 resampling level versus coefficient value for non-zero (absolute value > 1e-5) coefficients
-```{r, cache= FALSE, fig.keep="none"}
+```{r, cache= FALSE, fig.keep="none", eval = LOCAL}
 plot(Net_inf_C@network[abs(Net_inf_C@network)>1e-5],[email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 hist([email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
 Plot of confidence at .5 resampling level versus coefficient value for non-zero (absolute value > 1e-5) coefficients
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(Net_inf_C@network[abs(Net_inf_C@network)>1e-5],[email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 hist([email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
 Plot of confidence at .95 resamling level with groups created by thresholding the correlation matrix versus coefficient value for non-zero (absolute value > 1e-5) coefficients.
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(Net_inf_C@network[abs(Net_inf_C@network)>1e-5],[email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 hist([email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
 Plot of confidence at .95 resampling level versus coefficient value for non-zero (absolute value > 1e-5) coefficients using standard cross-validation.
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 plot(Net_inf_C@network[abs(Net_inf_C@network)>1e-5],[email protected][abs(Net_inf_C@network)>1e-5])
 ```
 
-```{r, cache= TRUE, fig.keep='none'}
+```{r, cache= TRUE, fig.keep='none', eval = LOCAL}
 hist([email protected][abs(Net_inf_C@network)>1e-5])
 ```