Merge branch 'hahmmr' into main

.circleci/config.yml

@@ -2,7 +2,7 @@ version: 2
 jobs:
   build:
     docker:
-      - image: rocker/verse:4.2
+      - image: rocker/verse:4.3
     environment:
             _R_CHECK_FORCE_SUGGESTS_: false
     steps:

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: numbat
 Title: Haplotype-Aware CNV Analysis from scRNA-Seq
 URL: https://github.com/kharchenkolab/numbat/, https://kharchenkolab.github.io/numbat/
-Version: 1.3.3
+Version: 1.4.0
 Authors@R: c(person("Teng","Gao", email="[email protected]", role=c("cre", "aut")), person("Ruslan", "Soldatov", email="[email protected]", role="aut"), person("Hirak", "Sarkar", email="[email protected]", role="aut"), person("Evan", "Biederstedt", email="[email protected]", role="aut"), person("Peter", "Kharchenko", email = "[email protected]", role = "aut"))
 Description: A computational method that infers copy number variations (CNVs) in cancer scRNA-seq data and reconstructs the tumor phylogeny. 'numbat' integrates signals from gene expression, allelic ratio, and population haplotype structures to accurately infer allele-specific CNVs in single cells and reconstruct their lineage relationship. 'numbat' can be used to: 1. detect allele-specific copy number variations from single-cells; 2. differentiate tumor versus normal cells in the tumor microenvironment; 3. infer the clonal architecture and evolutionary history of profiled tumors. 'numbat' does not require tumor/normal-paired DNA or genotype data, but operates solely on the donor scRNA-data data (for example, 10x Cell Ranger output). Additional examples and documentations are available at <https://kharchenkolab.github.io/numbat/>. For details on the method please see Gao et al. Nature Biotechnology (2022) <doi:10.1038/s41587-022-01468-y>.
 License: MIT + file LICENSE
@@ -21,6 +21,7 @@ Imports:
     ggraph,
     ggtree,
     glue,
+    hahmmr,
     igraph,
     IRanges,
     logger,

NAMESPACE

@@ -32,6 +32,16 @@ importFrom(data.table,fread)
 importFrom(data.table,fwrite)
 importFrom(ggtree,"%<+%")
 importFrom(grDevices,colorRampPalette)
+importFrom(hahmmr,dbbinom)
+importFrom(hahmmr,dpoilog)
+importFrom(hahmmr,fit_lnpois_cpp)
+importFrom(hahmmr,forward_back_allele)
+importFrom(hahmmr,l_bbinom)
+importFrom(hahmmr,l_lnpois)
+importFrom(hahmmr,likelihood_allele)
+importFrom(hahmmr,logSumExp)
+importFrom(hahmmr,run_allele_hmm_s5)
+importFrom(hahmmr,run_joint_hmm_s15)
 importFrom(igraph,"E<-")
 importFrom(igraph,"V<-")
 importFrom(igraph,E)
@@ -74,4 +84,4 @@ importFrom(stats,setNames)
 importFrom(stats,start)
 importFrom(stats,t.test)
 importFrom(utils,combn)
-useDynLib(numbat)
+useDynLib(numbat, .registration=TRUE)

NEWS.md

@@ -1,3 +1,9 @@
+# numbat 1.4.0 - 10/26/2023
+
+* Integration with hahmmr 
+
+* Better input checking for pileup_and_phase
+
 # numbat 1.3.3 - 08/15/2023
 
 * Fix plotting issue #135

R/RcppExports.R

@@ -1,43 +1,7 @@
 # Generated by using Rcpp::compileAttributes() -> do not edit by hand
 # Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
 
-cppdbbinom <- function(x, size, alpha, beta, log_prob = FALSE) {
-    .Call('_numbat_cppdbbinom', PACKAGE = 'numbat', x, size, alpha, beta, log_prob)
-}
-
-cpp_dgpois <- function(x, alpha, beta, log_prob = FALSE) {
-    .Call('_numbat_cpp_dgpois', PACKAGE = 'numbat', x, alpha, beta, log_prob)
-}
-
-logSumExp <- function(x) {
-    .Call('_numbat_logSumExp', PACKAGE = 'numbat', x)
-}
-
-likelihood_compute <- function(logphi, logprob, logPi, n, m) {
-    .Call('_numbat_likelihood_compute', PACKAGE = 'numbat', logphi, logprob, logPi, n, m)
-}
-
-forward_backward_compute <- function(logphi, logprob, logPi, n, m) {
-    .Call('_numbat_forward_backward_compute', PACKAGE = 'numbat', logphi, logprob, logPi, n, m)
-}
-
-viterbi_compute <- function(log_delta, logprob, logPi, n, m, nu, z) {
-    .Call('_numbat_viterbi_compute', PACKAGE = 'numbat', log_delta, logprob, logPi, n, m, nu, z)
-}
-
 roman2int_internal <- function(letters, nchar) {
-    .Call('_numbat_roman2int_internal', PACKAGE = 'numbat', letters, nchar)
-}
-
-fit_lnpois_cpp <- function(Y_obs, lambda_ref, d) {
-    .Call('_numbat_fit_lnpois_cpp', PACKAGE = 'numbat', Y_obs, lambda_ref, d)
-}
-
-poilog1 <- function(x, my, sig) {
-    .Call('_numbat_poilog1', PACKAGE = 'numbat', x, my, sig)
-}
-
-l_lnpois_cpp <- function(Y_obs, lambda_ref, d, mu, sig, phi = 1.0) {
-    .Call('_numbat_l_lnpois_cpp', PACKAGE = 'numbat', Y_obs, lambda_ref, d, mu, sig, phi)
+    .Call(`_numbat_roman2int_internal`, letters, nchar)
 }
 

R/diagnostics.R

@@ -1,6 +1,5 @@
 
 #' Log memory usage
-#' @param msg string Message to log
 #' @return NULL
 #' @keywords internal
 log_mem = function() {

R/genotyping.R

@@ -211,9 +211,6 @@ preprocess_allele = function(
 
     # phased VCF
     vcf_phased = vcf_phased %>% 
-        mutate(INFO = str_remove_all(INFO, '[:alpha:]|=')) %>%
-        tidyr::separate(col = 'INFO', into = c('AD', 'DP', 'OTH'), sep = ';') %>%
-        mutate_at(c('AD', 'DP', 'OTH'), as.integer) %>%
         mutate(snp_id = paste(CHROM, POS, REF, ALT, sep = '_')) %>%
         mutate(GT = get(sample))