Commit e85bf4bd authored by Lorenz Kapsner's avatar Lorenz Kapsner
Browse files

updated vignette and used custom glmpredict function for prediction of...

updated vignette and used custom glmpredict function for prediction of diagnostic signatures in validation data
parent 8bc25a56
......@@ -328,7 +328,9 @@ validateDiagnosticSignature_ <- function(validationstudylist,
for (j in c("min", "1se")){
predicted <- predict(models[[i]][[j]]$model, v.data.all, type = "response")
predicted <- predictGLM_(model = models[[i]][[j]]$model,
test.x = v.data.all,
type = "response")
confmat <- caret::confusionMatrix(data = factor(ifelse(as.numeric(as.character(predicted)) < 0.5, 0, 1)),
reference = factor(diagnosis),
......
......@@ -1367,7 +1367,7 @@ dir.create(csvdir)
species <- "Hs"
OrgDb <- "org.Hs.eg.db"
organism <- "hsa"
pathwayid <- "hsa04110"
pathwayid <- "hsa04151"
# diagnostig signature
seed <- 111
......@@ -1435,11 +1435,11 @@ sigident::sigidentEnrichment(mergeset = mergeset,
csvdir = csvdir)
```
```{r eval = FALSE}
filename <- paste0(plotdir, "/", organism, "04110.png")
filename <- paste0(plotdir, "/", pathwayid, ".png")
knitr::include_graphics(filename)
```
```{r eval = FALSE}
filename <- paste0(plotdir, "/", organism, "04110.pathview.png")
filename <- paste0(plotdir, "/", pathwayid, ".pathview.png")
knitr::include_graphics(filename)
```
......@@ -1466,6 +1466,31 @@ knitr::kable(
)
```
### Validation of the diagnostic signatures
Some users may validate the resulting models using new data sets. This can be conducted running the function `validateDiagnosticSignature_`. First, a list containing meta information on the validation study must be created.
```{r eval = FALSE}
validationstudylist <- list(studyname = "GSE30219",
setid = 1,
targetcolname = "source_name_ch1",
controllevelname = "Non Tumoral Lung",
targetlevelname = "Lung Tumour")
```
```{r eval = FALSE}
GSE30219_validation <- sigident::validateDiagnosticSignature_(
validationstudylist = validationstudylist,
models = diagnosticModels,
genes = genes,
idtype = idtype,
targetname = targetname,
controlname = controlname,
targetcol = targetcol,
datadir = datadir
)
```
## Run `sigidentPrognostic`-Function
### First, define meta information lists
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment