up vignette

DESCRIPTION

@@ -2,11 +2,15 @@ Package: scConform
 Type: Package
 Title: Conformal Inference For Cell Type Annotation
 Version: 0.99.0
-Authors@R: person("Corbetta", "Daniela", email = "daniela.corbetta@phd.unipd.it", role = c("aut", "cre"))
+Authors@R: person("Corbetta", "Daniela", 
+                  email = "daniela.corbetta@phd.unipd.it", 
+                  role = c("aut", "cre"))
 Maintainer: Daniela Corbetta <daniela.corbetta@phd.unipd.it>
-Description: Builds prediction interval for cell type annotation using conformal inference and conformal risk control.
-    It provides two main methods. The first one gives prediction intervals with coverage guarantees based 
-    on standard conformal inference. The second one instead gives hierarchical prediction intervals that are consistent
+Description: Builds prediction interval for cell type annotation using conformal 
+    inference and conformal risk control.
+    It provides two main methods. The first one gives prediction intervals with 
+    coverage guarantees based on standard conformal inference. The second one 
+    instead gives hierarchical prediction intervals that are consistent
     with the cell ontology.
 License: Artistic-2.0
 LazyData: true
@@ -18,12 +22,11 @@ Suggests:
     rmarkdown,
     BiocStyle,
     VGAM,
-    foreach,
     ontoProc,
     MerfishData,
     doParallel,
-    igraph,
-    scuttle
+    scuttle,
+    SingleCellExperiment
 Imports:
     foreach,
     igraph,

R/getPredictionSets.R

@@ -200,7 +200,7 @@ getPredictionSets <- function(x.query, x.cal, y.cal, onto = NULL, alpha = 0.1,
         # for(i in labels){
         #     p.query[,i] <- colData(x.query)[[i]]
         # }
-        p.query <- .retrievePredMatrix(x.query)
+        p.query <- .retrievePredMatrix(x.query, K=K)
     } else {
         p.query <- x.query
     }
@@ -212,7 +212,7 @@ getPredictionSets <- function(x.query, x.cal, y.cal, onto = NULL, alpha = 0.1,
         # for(i in labels){
         #     p.cal[,i] <- colData(x.cal)[[i]]
         # }
-        p.cal <- .retrievePredMatrix(x.cal)
+        p.cal <- .retrievePredMatrix(x.cal, K=K)
     } else {
         p.cal <- x.cal
     }
@@ -295,7 +295,7 @@ getPredictionSets <- function(x.query, x.cal, y.cal, onto = NULL, alpha = 0.1,
 ## function to retrieve prediction matrix from the colData of a
 ## SingleCellExperiment object
 
-.retrievePredMatrix <- function(sc) {
+.retrievePredMatrix <- function(sc, K) {
     n.sc <- ncol(sc)
     p.sc <- matrix(NA, nrow = n.sc, ncol = K)
     colnames(p.sc) <- labels

R/utils_HierarchicalPS.R

@@ -1,6 +1,7 @@
 # Function to get the hierarchical prediction sets for the observation in p.test
-# Needs a vector of lambda values. For each of the lambdas computes the prediction sets
-# for the data in the calibration set (p.cal n_cal x K matrix that contains
+# Needs a vector of lambda values. For each of the lambdas computes the
+# prediction sets for the data in the calibration set
+# (p.cal n_cal x K matrix that contains
 # estimated probabilities for each label). Based on these sets that compute the
 # loss table and then gets lambda hat based on equation (4) in Bates and
 # Angelopoulus (2023), Conformal Risk Control. Finally, builds prediction sets

vignettes/vignette.Rmd

@@ -24,7 +24,7 @@ knitr::opts_chunk$set(
 
 # Load useful packages
 
-```{r setup, message=FALSE}
+```{r libraries, message=FALSE}
 library(scConform)
 library(SingleCellExperiment)
 library(VGAM)
@@ -34,11 +34,38 @@ library(MerfishData)
 library(doParallel)
 library(igraph)
 library(scuttle)
-#
-# `%notin%` = Negate(`%in%`)
 
-# num_cores <- 8
-# cl1 <- makeCluster(num_cores)
-# registerDoParallel(cl1)
+`%notin%` = Negate(`%in%`)
+
+num_cores <- 2
+cl1 <- makeCluster(num_cores)
+registerDoParallel(cl1)
+```
+
+
+
+***
+
+## R.session Info
+
+```{r SessionInfo, echo=FALSE, message=FALSE, warning=FALSE, comment=NA}
+sessionInfo()
 ```
 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+