From 69ba7ee1e9db5a9e8b0a0fc6ba84f2eb4a4a01be Mon Sep 17 00:00:00 2001
From: Luke Zappia <lazappi@users.noreply.github.com>
Date: Thu, 22 Aug 2019 15:08:15 +1000
Subject: [PATCH] Add density sampling option for means, lib sizes
---
DESCRIPTION | 4 +--
NEWS.md | 4 +++
R/AllClasses.R | 18 +++++++++++-
R/SplotchParams-methods.R | 20 ++++++++++----
R/splotch-estimate.R | 8 ++++--
R/splotch-simulate.R | 58 +++++++++++++++++++++++++--------------
man/SplotchParams.Rd | 11 +++++++-
7 files changed, 92 insertions(+), 31 deletions(-)
diff --git a/DESCRIPTION b/DESCRIPTION
index b036793..1b17759 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,8 +1,8 @@
Package: splatter
Type: Package
Title: Simple Simulation of Single-cell RNA Sequencing Data
-Version: 1.9.4.9009
-Date: 2019-08-21
+Version: 1.9.4.9010
+Date: 2019-08-22
Author: Luke Zappia
Authors@R:
c(person("Luke", "Zappia", role = c("aut", "cre"),
diff --git a/NEWS.md b/NEWS.md
index 8ccfac0..99438d9 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,3 +1,7 @@
+### Version 1.9.4.9010 (2019-08-22)
+
+* Add density sampling options for means and library sizes
+
### Version 1.9.4.9009 (2019-08-21)
* Replace library size log-normal with density and rejection sampling
diff --git a/R/AllClasses.R b/R/AllClasses.R
index 3af99fe..f14b964 100644
--- a/R/AllClasses.R
+++ b/R/AllClasses.R
@@ -279,7 +279,13 @@ setClass("SplatParams",
#' the expression outlier factor log-normal distribution.}
#' \item{\code{mean.outFacScale}}{Scale (sdlog) parameter for the
#' expression outlier factor log-normal distribution.}
-#' \item{\code{mean.values}}{Vector of means for each gene.}
+#' \item{\code{mean.dens}}{\code{\link{density}} object describing
+#' the log gene mean density.}
+#' \item{\code{[mean.method]}}{Method to use for simulating gene
+#' means. Either "fit" to sample from a gamma distribution (with
+#' expression outliers) or "density" to sample from the provided
+#' density object.}
+#' \item{\code{[mean.values]}}{Vector of means for each gene.}
#' }
#' }
#' \item{\emph{Network parameters}}{
@@ -306,6 +312,9 @@ setClass("SplatParams",
#' distribution is used.}
#' \item{\code{lib.dens}}{\code{\link{density}} object describing
#' the library size density.}
+#' \item{\code{[lib.method]}}{Method to use for simulating library
+#' sizes. Either "fit" to sample from a log-normal distribution or
+#' "density" to sample from the provided density object.}
#' }
#' }
#' \item{\emph{Paths parameters}}{
@@ -331,6 +340,8 @@ setClass("SplotchParams",
mean.outProb = "numeric",
mean.outLoc = "numeric",
mean.outScale = "numeric",
+ mean.dens = "density",
+ mean.method = "character",
mean.values = "numeric",
network.graph = "ANY",
network.nRegs = "numeric",
@@ -341,12 +352,16 @@ setClass("SplotchParams",
lib.loc = "numeric",
lib.scale = "numeric",
lib.dens = "density",
+ lib.method = "character",
cells.design = "data.frame"),
prototype = prototype(mean.rate = 0.3,
mean.shape = 0.6,
mean.outProb = 0.05,
mean.outLoc = 4,
mean.outScale = 0.5,
+ mean.dens = density(rgamma(10000, rate = 0.3,
+ shape = 0.6)),
+ mean.method = "fit",
mean.values = numeric(),
network.graph = NULL,
network.nRegs = 100,
@@ -361,6 +376,7 @@ setClass("SplotchParams",
lib.loc = 11,
lib.scale = 0.2,
lib.dens = density(rlnorm(10000, 11, 0.2)),
+ lib.method = "fit",
cells.design = data.frame(
Path = 1,
Probability = 1,
diff --git a/R/SplotchParams-methods.R b/R/SplotchParams-methods.R
index 6f4d13b..56a1b6c 100644
--- a/R/SplotchParams-methods.R
+++ b/R/SplotchParams-methods.R
@@ -25,6 +25,9 @@ setValidity("SplotchParams", function(object) {
mean.outProb = checkNumber(v$mean.outProb, lower = 0, upper = 1),
mean.outLoc = checkNumber(v$mean.outLoc),
mean.outScale = checkNumber(v$mean.outScale, lower = 0),
+ mean.dens = checkmate::checkClass(v$mean.dens, "density"),
+ mean.method = checkmate::checkChoice(v$mean.method,
+ c("fit", "density")),
network.graph = checkmate::checkClass(v$network, "igraph",
null.ok = TRUE),
network.nRegs = checkmate::checkInt(v$network.nRegs,
@@ -40,6 +43,8 @@ setValidity("SplotchParams", function(object) {
lib.loc = checkmate::checkNumber(v$lib.loc),
lib.scale = checkmate::checkNumber(v$lib.scale, lower = 0),
lib.dens = checkmate::checkClass(v$lib.dens, "density"),
+ lib.method = checkmate::checkChoice(v$lib.method,
+ c("fit", "density")),
cells.design = checkmate::checkDataFrame(v$cells.design,
types = "numeric",
any.missing = FALSE,
@@ -124,6 +129,8 @@ setMethod("show", "SplotchParams", function(object) {
"(Out Prob)" = "mean.outProb",
"(Out Location)" = "mean.outLoc",
"(Out Scale)" = "mean.outScale",
+ "(Density)" = "mean.density",
+ "[Method]" = "mean.method",
"[Values]*" = "mean.values"))
pp.network <- list("Network:" = c("[Graph]" = "network.graph",
@@ -135,7 +142,8 @@ setMethod("show", "SplotchParams", function(object) {
pp.bot <- list("Library size:" = c("(Location)" = "lib.loc",
"(Scale)" = "lib.scale",
- "(Density)" = "lib.dens"),
+ "(Density)" = "lib.dens",
+ "[Method]" = "lib.method"),
"Cells:" = c("[Design]" = "cells.design"))
paths.means <- getParam(object, "paths.means")
@@ -205,10 +213,12 @@ setMethod("setParam", "SplotchParams", function(object, name, value) {
if (name == "network.graph") {
checkmate::assertClass(value, "igraph")
- object <- setParamUnchecked(object, "nGenes", igraph::gorder(value))
- if (!(length(getParam(object, "mean.values")) == 0)) {
- warning("changing network.graph resets mean.values")
- object <- setParam(object, "mean.values", numeric())
+ if (getParam(object, "nGenes") != igraph::gorder(value)) {
+ if (!(length(getParam(object, "mean.values")) == 0)) {
+ warning("changing network.graph resets mean.values")
+ object <- setParam(object, "mean.values", numeric())
+ }
+ object <- setParamUnchecked(object, "nGenes", igraph::gorder(value))
}
if ("IsReg" %in% igraph::vertex_attr_names(value)) {
diff --git a/R/splotch-estimate.R b/R/splotch-estimate.R
index 523e07b..cae0f88 100644
--- a/R/splotch-estimate.R
+++ b/R/splotch-estimate.R
@@ -78,7 +78,7 @@ splotchEstMean <- function(norm.counts, params, verbose) {
med <- median(lmeans)
mad <- mad(lmeans)
- bound <- med + 1 * mad
+ bound <- med + 2 * mad
outs <- which(lmeans > bound)
@@ -88,13 +88,16 @@ splotchEstMean <- function(norm.counts, params, verbose) {
if (length(outs) > 1) {
facs <- means[outs] / median(means)
- fit <- selectFit(facs, "lnorm", verbose = verbose)
+ #fit <- selectFit(facs, "lnorm", verbose = verbose)
+ fit <- fitdistrplus::fitdist(facs, "lnorm")
params <- setParams(params,
mean.outLoc = unname(fit$estimate["meanlog"]),
mean.outScale = unname(fit$estimate["sdlog"]))
}
+ params <- setParams(params, mean.dens = density(lmeans))
+
return(params)
}
@@ -191,6 +194,7 @@ selectFit <- function(data, distr, weights = NULL, verbose = TRUE) {
}
scores <- fitdistrplus::gofstat(fits)$cvm
+
# Flatten in case scores is a list
scores.flat <- unlist(scores)
selected <- which(scores.flat == min(scores.flat, na.rm = TRUE))
diff --git a/R/splotch-simulate.R b/R/splotch-simulate.R
index 1148c61..5632389 100644
--- a/R/splotch-simulate.R
+++ b/R/splotch-simulate.R
@@ -146,20 +146,30 @@ splotchSimGeneMeans <- function(params, verbose) {
if (verbose) {message("Simulating means...")}
nGenes <- getParam(params, "nGenes")
- mean.shape <- getParam(params, "mean.shape")
- mean.rate <- getParam(params, "mean.rate")
- mean.outProb <- getParam(params, "mean.outProb")
- mean.outLoc <- getParam(params, "mean.outLoc")
- mean.outScale <- getParam(params, "mean.outScale")
-
- mean.values <- rgamma(nGenes, shape = mean.shape, rate = mean.rate)
-
- outlier.facs <- getLNormFactors(nGenes, mean.outProb, 0, mean.outLoc,
- mean.outScale)
- median.means.gene <- median(mean.values)
- outlier.means <- median.means.gene * outlier.facs
- is.outlier <- outlier.facs != 1
- mean.values[is.outlier] <- outlier.means[is.outlier]
+ mean.method <- getParam(params, "mean.method")
+
+ if (mean.method == "fit") {
+ if (verbose) {message("Sampling from gamma distribution...")}
+ mean.shape <- getParam(params, "mean.shape")
+ mean.rate <- getParam(params, "mean.rate")
+ mean.outProb <- getParam(params, "mean.outProb")
+ mean.outLoc <- getParam(params, "mean.outLoc")
+ mean.outScale <- getParam(params, "mean.outScale")
+
+ mean.values <- rgamma(nGenes, shape = mean.shape, rate = mean.rate)
+
+ outlier.facs <- getLNormFactors(nGenes, mean.outProb, 0, mean.outLoc,
+ mean.outScale)
+ median.means.gene <- median(mean.values)
+ outlier.means <- median.means.gene * outlier.facs
+ is.outlier <- outlier.facs != 1
+ mean.values[is.outlier] <- outlier.means[is.outlier]
+ } else if (mean.method == "density") {
+ if (verbose) {message("Sampling from density object...")}
+ mean.dens <- getParam(params, "mean.dens")
+
+ mean.values <- exp(sampleDensity(nGenes, mean.dens, lower = -Inf))
+ }
params <- setParam(params, "mean.values", mean.values)
@@ -242,12 +252,20 @@ splotchSimLibSizes <- function(sim, params, verbose) {
if (verbose) {message("Simulating library sizes...")}
nCells <- getParam(params, "nCells")
- # lib.loc <- getParam(params, "lib.loc")
- # lib.scale <- getParam(params, "lib.scale")
- lib.dens <- getParam(params, "lib.dens")
+ lib.method <- getParam(params, "lib.method")
- # exp.lib.sizes <- rlnorm(nCells, lib.loc, lib.scale)
- exp.lib.sizes <- rejectionSample(nCells, lib.dens)
+ if (lib.method == "fit") {
+ if (verbose) {message("Sampling from log-normal distribution...")}
+ lib.loc <- getParam(params, "lib.loc")
+ lib.scale <- getParam(params, "lib.scale")
+
+ exp.lib.sizes <- rlnorm(nCells, lib.loc, lib.scale)
+ } else if (lib.method == "density") {
+ if (verbose) {message("Sampling from density object...")}
+ lib.dens <- getParam(params, "lib.dens")
+
+ exp.lib.sizes <- sampleDensity(nCells, lib.dens)
+ }
colData(sim)$ExpLibSize <- exp.lib.sizes
@@ -352,7 +370,7 @@ getBetaStepProbs <- function(steps, alpha, beta) {
}
#' @importFrom stats approxfun
-rejectionSample <- function(n, dens, lower = 0) {
+sampleDensity <- function(n, dens, lower = 0) {
xmin <- min(dens$x)
xmax <- max(dens$x)
diff --git a/man/SplotchParams.Rd b/man/SplotchParams.Rd
index 2c7861c..0b03061 100644
--- a/man/SplotchParams.Rd
+++ b/man/SplotchParams.Rd
@@ -29,7 +29,13 @@ The Splotch simulation uses the following parameters:
the expression outlier factor log-normal distribution.}
\item{\code{mean.outFacScale}}{Scale (sdlog) parameter for the
expression outlier factor log-normal distribution.}
- \item{\code{mean.values}}{Vector of means for each gene.}
+ \item{\code{mean.dens}}{\code{\link{density}} object describing
+ the log gene mean density.}
+ \item{\code{[mean.method]}}{Method to use for simulating gene
+ means. Either "fit" to sample from a gamma distribution (with
+ expression outliers) or "density" to sample from the provided
+ density object.}
+ \item{\code{[mean.values]}}{Vector of means for each gene.}
}
}
\item{\emph{Network parameters}}{
@@ -56,6 +62,9 @@ The Splotch simulation uses the following parameters:
distribution is used.}
\item{\code{lib.dens}}{\code{\link{density}} object describing
the library size density.}
+ \item{\code{[lib.method]}}{Method to use for simulating library
+ sizes. Either "fit" to sample from a log-normal distribution or
+ "density" to sample from the provided density object.}
}
}
\item{\emph{Paths parameters}}{
--
GitLab