Version 1.0.6: updated EPIC based on the latest reference profiles, g…

…iving them new names. Documentation has also been updated.

DESCRIPTION

@@ -1,16 +1,17 @@
 Package: EPIC
 Type: Package
 Title: Estimate the Proportion of Immune and Cancer cells
-Version: 1.0.5
+Version: 1.0.6
 Authors@R: as.person(c(
   "Julien Racle <[email protected]> [aut, cre]",
   "David Gfeller <[email protected]> [aut]"
   ))
 Description: Package implementing EPIC method to estimate the proportion of
-    immune and cancer cells from bulk gene expression data. It is based on
-    reference gene expression profiles for the main immune cell types and it
-    predicts the proportion of these cells and of the remaining "other cells"
-    that are the remaining cells (cancer, stromal, endothelial) for which no
+    immune, stromal, endothelial and cancer or other cells from bulk gene
+    expression data.
+    It is based on reference gene expression profiles for the main non-malignant
+    cell types and it predicts the proportion of these cells and of the
+    remaining "other cells" (that are mostly cancer cells) for which no
     reference profile is given.
 Depends: R (>= 3.2.0)
 License: file LICENSE

NEWS

@@ -1,3 +1,10 @@
+Version 1.0.6
+------------------------------------------------------------------------
+* Updated EPIC to use the latest reference profiles and updated some default
+  values and tests for the package.
+* Updated the help files accordingly, explaining also the additional advanced
+  options available for EPIC.
+
 Version 1.0.5
 ------------------------------------------------------------------------
 * Updated EPIC to remove genes from bulk that have only NA values.
@@ -11,7 +18,7 @@ Version 1.0.4
   release the constrain that the sum of cells with known reference profiles
   must be smaller than 1. Also if withOtherCells is FALSE and constrainedSum is
   FALSE, we don't use the constraint during the optimization that the sum of all
-  the cells must also be equal to 0 (note we still rescale afterwards the
+  the cells must also be equal to 1 (note we still rescale afterwards the
   proportions to have a sum equal to 1 but it isn't done during the
   optimization).
 
@@ -40,8 +47,8 @@ Version 1.0.1
   the p-value isn't exact in the presence of ties.
 * Added mRNA_cell values for T cell subtypes.
 * Added reference profiles including CD4 and CD8 T cells for circulating
-  immune cells (BRef_s, BRef_s.tpm) and including also CAFs and endothelial
-  cells for tumor infilatring cells (TRef_s.tpm).
+  immune cells and including also CAFs and endothelial cells for tumor
+  infilatring cells.
 
 
 Version 1.0.0

R/EPIC_descr.R

@@ -1,23 +1,22 @@
 #' EPIC: a package to Estimate the Proportion of Immune and Cancer cells from
 #'  tumor gene expression data.
 #'
-#' EPIC package provides the function and immune cell reference profiles to
-#' estimate the proportion of immune and other cells from bulk gene expression
-#' data.
+#' EPIC package provides the function and cell reference profiles to
+#' estimate the proportion of immune, stromal, endothelial and cancer or other
+#' cells from bulk gene expression data.
 #'
 #' @section EPIC functions:
 #' \code{\link{EPIC}} is the main function to call to estimate the
 #'  various cells proportions from a bulk sample.
 #'
 #' @section Included datasets:
-#' \code{\link{BRef}}, \code{\link{BRef.tpm}}: reference profiles from
-#'    circulating immune cells.
+#' \code{\link{BRef}}: reference profiles from circulating immune cells.
 #'
-#'  \code{\link{TRef.tpm}}: reference profiles from immune cells obtained
-#'  from single cell data of tumor infiltrating cells from melanoma patients.
+#'  \code{\link{TRef}}: reference profiles from tumor infiltrating non-malignant
+#'  cells obtained from single cell data of melanoma patients.
 #'
-#' \code{\link{hoek_data}}: example dataset containing data from Hoek et al,
-#'  2015, PLoS One.
+#' \code{\link{melanoma_data}}: example dataset containing data from lymph nodes
+#'  from patients with metastatic melanoma.
 #'
 #' \code{\link{mRNA_cell_default}}: values of mRNA per cell for the main cell
 #'  types.
@@ -34,20 +33,17 @@ NULL
 #' Reference profiles from circulating immune cells.
 #'
 #' A dataset containing the reference profiles obtained from immune cell
-#' samples of \emph{B-}, \emph{NK-}, \emph{T-cells}, \emph{Monocytes}
-#' and \emph{Neutrophils} purified from PBMC or whole blood.
+#' samples of \emph{B cells}, \emph{CD4 T cells}, \emph{CD8 T cells},
+#' \emph{Monocytes}, \emph{NK cells} and \emph{Neutrophils}, purified from
+#' PBMC or whole blood.
 #'
 #' The original samples were obtained from healthy donors and donors after
 #' influenza vaccination or with diabetes, sepsis or multiple sclerosis.
 #'
-#' @section Similar datasets:
-#'  \code{BRef} (main reference profiles, using data from sources 1-3 below)
-#'
-#'
 #' @format A list of 3 elements: \describe{ \item{$refProfiles,
-#'   $refProfiles.var}{Matrices (nGenes x nRefCells) of the normalized gene
-#'   expression from the reference cells and the variability of this gene
-#'   expression for each gene and each cell type} \item{$sigGenes}{A list of 100
+#'   $refProfiles.var}{Matrices (nGenes x nRefCells) of the gene expression (in
+#'   TPM counts) from the reference cells and the variability of this gene
+#'   expression for each gene and each cell type} \item{$sigGenes}{A list of
 #'   signature genes used to deconvolve the cell proportions} }
 #'
 #' @source \enumerate{
@@ -57,32 +53,32 @@ NULL
 #'  }
 "BRef"
 
-#' @section Similar datasets:
-#' \code{BRef.tpm} (reference profiles based on same data as \code{BRef},
-#'  but given in TPM counts instead of raw counts)
-#' @rdname BRef
-"BRef.tpm"
 
 #' Reference profiles obtained from single cell data of tumor infiltrating
 #' cells.
 #'
-#' A dataset containing the reference profiles given in TPM from various cell
-#' types: \emph{B-}, \emph{NK-}, \emph{T-cells} and \emph{Macrophages}.
+#' A dataset containing the reference profiles obtained from various
+#' tumor infiltrating non-malignant cell types: \emph{B cells},
+#' \emph{cancer-associated fibroblasts}, \emph{CD4 T cells}, \emph{CD8 T cells},
+#' \emph{endothelial cells}, \emph{Macrophages} and \emph{NK cells}.
 #'
 #' These were obtained from single-cell RNA-seq data from 9 donors from
-#' the publication of Tirosh et al., 2016, Science. The samples
-#' come from cancer metastases of melanoma (extracted from primary tumors
-#' and non-lymphoid tissue metastases). The classification for each sample with
-#' respect to each cell type is the one given by Tirosh et al.
+#' the publication of \cite{Tirosh et al., 2016, Science}. The samples
+#' come from melanoma tumors (extracted from primary tumors and non-lymphoid
+#' tissue metastases). The classification for each sample with
+#' respect to each cell type is the one given by Tirosh et al., except for
+#' the CD4 T cells and CD8 T cells, that were identified from the T cells based
+#' on the expression of CD4, CD8A and CD8B as described in \cite{EPIC}
+#' publication.
 #'
 #' @format A list of 3 elements: \describe{ \item{$refProfiles,
-#'   $refProfiles.var}{Matrices (nGenes x nRefCells) of the normalized gene
-#'   expression from the reference cells and the variability of this gene
-#'   expression for each gene and each cell type} \item{$sigGenes}{A list of 80
+#'   $refProfiles.var}{Matrices (nGenes x nRefCells) of the gene expression (in
+#'   TPM counts) from the reference cells and the variability of this gene
+#'   expression for each gene and each cell type} \item{$sigGenes}{A list of
 #'   signature genes used to deconvolve the cell proportions} }
 #'
 #' @source \url{http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE72056}
-"TRef.tpm"
+"TRef"
 
 #' Values of mRNA / cell for the main cell types.
 #'
@@ -99,28 +95,32 @@ NULL
 #'  \code{mRNA_cell_default} vector.
 "mRNA_cell_default"
 
-#' Example dataset containing data from Hoek et al, 2015, PLoS One.
+#' Example dataset containing data from lymph nodes from patients with
+#' metastatic melanoma.
 #'
-#' This dataset contains a subset of the full Hoek et al data. It contains only
-#' the data from the two healthy donors PBMC before influenza vaccination.
+#' This is the dataset obtained for \cite{EPIC} publication. It contains the
+#' gene expression from lymph node samples from four patients with melanoma, and
+#' it contains also the proportions of the main immune cell types and of
+#' melanoma cells, as measured by FACS.
 #'
 #' @format This is a list of 3 elements: \describe{
-#'  \item{$rawCounts}{(matrix of 51574 genes x 2 donors) The raw read counts
-#'    from the two donors. It has been obtained by mapping the original
-#'    fastq files to \emph{hg19} genome with help of \emph{tophat} and
-#'    \emph{htseq-count}.}
-#'  \item{$cellFractions.obs}{(matrix of 2 donors x 6 cell types) The
-#'    proportions of the different immune cells in the PBMC from the 2 donors,
-#'    as measured by FACS by Hoek et al.}
-#'  \item{$cellFractions.pred}{(matrix of 2 donors x 7 cell types) The
-#'    proportions of the different immune cells and of a potential additional
-#'    uncharacterized cell, as predicted by EPIC based on the rawCounts and
-#'    the profiles \code{reference=BRef}.}
+#'  \item{$counts}{(matrix of 49902 genes x 4 donors) The TPM normalized counts
+#'    from the four donors. It has been obtained by mapping RNA-seq data to
+#'    \emph{hg19} genome with help of \emph{RSEM}. Ensembl ID were then
+#'    converted to gene names, and genes with duplicated entries were
+#'    merged together by summing their counts.}
+#'  \item{$cellFractions.obs}{(matrix of 4 donors x 6 cell types) The
+#'    proportions of the different cell types measured by FACS (the
+#'    "other_cells" correspond to the live cells without any marker of the
+#'    other given cell types).}
+#'  \item{$cellFractions.pred}{(matrix of 4 donors x 8 cell types) The
+#'    proportions of the different cell types, as predicted by EPIC based on
+#'    the reference profiles \code{TRef}.}
 #' }
 #'
 #' @source The description of this data can be found here:
-#'  \href{http://dx.doi.org/10.1371/journal.pone.0118528}{link to paper}
-#'  and \href{https://www.ebi.ac.uk/arrayexpress/experiments/E-GEOD-64655}{link
+#'  \href{http://www.biorxiv.org/content/early/2017/03/17/117788}{link to paper}
+#'  and \href{https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE93722}{link
 #'  to data}.
-"hoek_data"
+"melanoma_data"