From 1cbc480fe6efb035a0ad6cd5ad407403a16935f0 Mon Sep 17 00:00:00 2001
From: marcjwilliams1 <marcjwilliams1@gmail.com>
Date: Fri, 3 Sep 2021 11:26:41 -0400
Subject: [PATCH] bugfix: specify filtering in ascn

---
 R/callASCN.R                   | 5 +++--
 R/callHSCN.R                   | 2 +-
 man/callAlleleSpecificCN.Rd    | 4 +++-
 man/callHaplotypeSpecificCN.Rd | 4 +++-
 4 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/R/callASCN.R b/R/callASCN.R
index fff3a57..ede323b 100644
--- a/R/callASCN.R
+++ b/R/callASCN.R
@@ -177,6 +177,7 @@ switch_alleles <- function(cn) {
 #' @param minbins Minimum number of bins containing both haplotype counts and copy number data for a cell to be included
 #' @param minbinschr Minimum number of bins containing both haplotype counts and copy number data per chromosome for a cell to be included
 #' @param maxloherror Maximum value for LOH error rate
+#' @param filterhaplotypes filter out haplotypes present in less than X fraction, default is 0.1
 #'
 #' @return allele specific copy number object which includes dataframe similar to input with additional columns which include
 #'
@@ -204,7 +205,7 @@ callAlleleSpecificCN <- function(CNbins,
                                  minbins = 100,
                                  minbinschr = 10,
                                  maxloherror = 0.03,
-                                 filterhaplotypes = TRUE) {
+                                 filterhaplotypes = 0.1) {
   if (!likelihood %in% c("binomial", "betabinomial", "auto")) {
     stop("Likelihood model for HMM emission model must
          be one of binomial, betabinomial or auto",
@@ -226,7 +227,7 @@ callAlleleSpecificCN <- function(CNbins,
   }
   
   if (filterhaplotypes){
-    haplotypes <- filter_haplotypes(haplotypes)
+    haplotypes <- filter_haplotypes(haplotypes, filterhaplotypes)
   }
 
   CNBAF <- combineBAFCN(haplotypes = haplotypes, CNbins = CNbins)
diff --git a/R/callHSCN.R b/R/callHSCN.R
index 157a0d0..4b97ac2 100644
--- a/R/callHSCN.R
+++ b/R/callHSCN.R
@@ -549,7 +549,7 @@ filter_haplotypes <- function(haplotypes, fraction){
   haplotypes <- as.data.table(haplotypes)
   nhaps <- dim(haplotypes)[1]
   total_cells <- length(unique(haplotypes$cell_id))
-  message("Filtering out haplotypes present < 10% of cells...")
+  message(paste0("Filtering out haplotypes present < ", fraction * 100, "% of cells..."))
   haplotypes <- haplotypes %>% 
     .[, ncells := length(unique(cell_id)), by = .(chr, start, end, hap_label)] %>% 
     .[, fcells := ncells / total_cells] %>% 
diff --git a/man/callAlleleSpecificCN.Rd b/man/callAlleleSpecificCN.Rd
index 7d7bffb..866cce6 100644
--- a/man/callAlleleSpecificCN.Rd
+++ b/man/callAlleleSpecificCN.Rd
@@ -17,7 +17,7 @@ callAlleleSpecificCN(
   minbins = 100,
   minbinschr = 10,
   maxloherror = 0.03,
-  filterhaplotypes = TRUE
+  filterhaplotypes = 0.1
 )
 }
 \arguments{
@@ -44,6 +44,8 @@ callAlleleSpecificCN(
 \item{minbinschr}{Minimum number of bins containing both haplotype counts and copy number data per chromosome for a cell to be included}
 
 \item{maxloherror}{Maximum value for LOH error rate}
+
+\item{filterhaplotypes}{filter out haplotypes present in less than X fraction, default is 0.1}
 }
 \value{
 allele specific copy number object which includes dataframe similar to input with additional columns which include
diff --git a/man/callHaplotypeSpecificCN.Rd b/man/callHaplotypeSpecificCN.Rd
index 57f9449..4028bf1 100644
--- a/man/callHaplotypeSpecificCN.Rd
+++ b/man/callHaplotypeSpecificCN.Rd
@@ -25,7 +25,7 @@ callHaplotypeSpecificCN(
   overwritemincells = NULL,
   cluster_per_chr = TRUE,
   viterbiver = "cpp",
-  filterhaplotypes = TRUE
+  filterhaplotypes = 0.1
 )
 }
 \arguments{
@@ -64,6 +64,8 @@ callHaplotypeSpecificCN(
 \item{overwritemincells}{Overwrite the the number of cells to use for clustering/phasing}
 
 \item{cluster_per_chr}{Whether to cluster per chromosome to rephase alleles or not}
+
+\item{filterhaplotypes}{filter out haplotypes present in less than X fraction, default is 0.1}
 }
 \value{
 allele specific copy number object which includes dataframe similar to input with additional columns which include