version 1.1.3

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: g3viz
 Type: Package
 Title: Interactively Visualize Genetic Mutation Data using a Lollipop-Diagram
-Version: 1.1.2
+Version: 1.1.3
 Authors@R: c(
     person("Xin", "Guo", email = "[email protected]", role = c("aut", "cre"))
     )
@@ -25,5 +25,5 @@ Suggests:
     kableExtra
 URL: https://github.com/G3viz/g3viz
 BugReports: https://github.com/G3viz/g3viz/issues
-RoxygenNote: 6.1.1
+RoxygenNote: 7.1.1
 VignetteBuilder: knitr

NAMESPACE

@@ -10,6 +10,7 @@ export(hgnc2pfam)
 export(hgnc2uniprot)
 export(mapMutationTypeToMutationClass)
 export(parseProteinChange)
+export(proteinDomain)
 export(readMAF)
 export(renderG3Lollipop)
 export(uniprot2pfam)

R/g3Lollipop.R

@@ -62,6 +62,9 @@ g3Lollipop <- function(mutation.dat,
                        # legend factor
                        factor.col = "Mutation_Class",
                        plot.options = g3Lollipop.options(),
+                       # ---------------------
+                       domain.info = NULL,
+                       # ---------------------
                        save.png.btn = TRUE,
                        save.svg.btn = TRUE,
                        btn.style = NA,
@@ -96,23 +99,15 @@ g3Lollipop <- function(mutation.dat,
     stop("Some columns are missing in mutation data: ", paste(missing.columns, collapse = ", "))
   }
 
-  # get mutation data for the given gene
-  snv.data.df <- mutation.dat[mutation.dat[, gene.symbol.col] == gene.symbol
-                              & !is.na(mutation.dat[, aa.pos.col]), ]
-  snv.data.json <- toJSON(snv.data.df, pretty = FALSE, auto_unbox = TRUE)
-
+  # =======================================
+  # version 1.2.0
+  # (1) custom domain information
+  #     Require: domain
+  # (2) custom domain information format
+  #     Require
   # get protein domain information
   domain.data.json <- hgnc2pfam(hgnc.symbol = gene.symbol,
-																uniprot.id = uniprot.id)
-
-  # read in data
-  snv.data.format <- list(
-    x = aa.pos.col,
-    y = protein.change.col,
-    factor = factor.col
-  )
-
-  snv.data.format.json <- toJSON(snv.data.format, pretty = FALSE, auto_unbox = TRUE)
+                                uniprot.id = uniprot.id)
 
   # domain data format
   domain.data.format <- list(
@@ -125,6 +120,21 @@ g3Lollipop <- function(mutation.dat,
     )
   )
   domain.data.format.json <- toJSON(domain.data.format, pretty = FALSE, auto_unbox = TRUE)
+  # =======================================
+
+  # get mutation data for the given gene
+  snv.data.df <- mutation.dat[mutation.dat[, gene.symbol.col] == gene.symbol
+                              & !is.na(mutation.dat[, aa.pos.col]), ]
+  snv.data.json <- toJSON(snv.data.df, pretty = FALSE, auto_unbox = TRUE)
+
+  # read in data
+  snv.data.format <- list(
+    x = aa.pos.col,
+    y = protein.change.col,
+    factor = factor.col
+  )
+
+  snv.data.format.json <- toJSON(snv.data.format, pretty = FALSE, auto_unbox = TRUE)
 
   plot.options.json <- toJSON(plot.options, pretty = FALSE, auto_unbox = TRUE)
 

R/proteinDomain.R

@@ -0,0 +1,84 @@
+#' Map from Hugo symbol to Pfam domains
+#'
+#' @description Mapping from Hugo symbol to Pfam-A domain composition.
+#' If the given Hugo symbol has multiple UniProt ID mappings,
+#' and \code{guess == TRUE},
+#' the longest UniProt protein is selected. Return is either a list of a JSON.
+#' @examples
+#' # general usage
+#' hgnc2pfam("TP53")
+#' hgnc2pfam("TP53", output.format = "json")
+#' hgnc2pfam("TP53", output.format = "list")
+#' hgnc2pfam("TP53", output.format = "json", uniprot.id = "P04637") # OK
+#'
+#' # for gene mapping to multiple UniProt enties
+#' hgnc2pfam("GNAS", guess = TRUE)
+#' hgnc2pfam("GNAS", guess = FALSE)
+#' hgnc2pfam("GNAS", output.format = "list")
+#' hgnc2pfam("GNAS", output.format = "list", uniprot.id = "P84996")
+#' \dontrun{
+#' hgnc2pfam("GNAS", output.format = "list", uniprot.id = "P84997") # not exists, returns FALSE
+#' }
+#'
+#' hgnc2pfam("PRAMEF9", output.format = "list") # no Pfam mappings
+#'
+#' @param hgnc.symbol primary Hugo symbol
+#' @param output.format output format: JSON or list
+#' @param uniprot.id UniProt ID, in case that gene symbol maps to multiple UniProt entries.
+#' @param guess if the given Hugo symbol links to multiple UniProt IDs,
+#' choose the longest one (\code{guess == TRUE});
+#' otherwise \code{NA} (\code{guess == FALSE}).  Default \code{TRUE}.
+#' @return A list or a JSON with attributes:
+#'     \emph{symbol}, \emph{uniprot}, \emph{length}, and a list of \emph{Pfam} entries, including
+#'     \emph{hmm.acc}, \emph{hmm.name}, \emph{start}, \emph{end}, and \emph{type}.
+#'
+#' @importFrom utils capture.output
+#' @importFrom jsonlite toJSON
+#' @export
+proteinDomain <- function(protein.length,
+                          domain.df,
+                          domain.name.col = "Name",
+                          domain.start.col = "Start",
+                          domain.end.col = "End"){
+
+  if(is.na(protein.length)){
+    stop("Please provide the length of protein")
+  }
+
+  if(!domain.name.col %in% colnames(domain.df)){
+    stop("domain.name.col ", domain.name.col, " not exists")
+  }
+  if(!domain.start.col %in% colnames(domain.df)){
+    stop("domain.start.col ", domain.start.col, " not exists")
+  }
+  if(!domain.end.col %in% colnames(domain.df)){
+    stop("domain.end.col ", domain.end.col, " not exists")
+  }
+
+  #--
+  protein.length <- 393
+  domain.fn = "data-raw/TP53_P04637_interpro.csv"
+  domain.df = read.csv(domain.fn)
+  domain.name.col = "Name"
+  domain.start.col = "Start"
+  domain.end.col = "End"
+  #--
+
+
+
+  domain.list <- list(
+    length = protein.length,
+    details = domain.list
+  )
+
+  # domain data format
+  domain.default.format <- list(
+    length = "length",
+    details = list(
+      start = "start",
+      end = "end",
+      name = "name"
+    )
+  )
+
+}

R/zzz.R

@@ -1,3 +1,3 @@
 .onAttach <- function(libname, pkgname){
-  packageStartupMessage("g3viz: visualizing gene/genome/gentics data for fun.\nAny questions, please send emails to <[email protected]>")
+  packageStartupMessage("g3viz: visualizing gene/genome/gentics data for fun.\nAny questions, please send emails to <[email protected]> or post on GitHub <https://github.com/G3viz/g3viz/issues>.")
 }

README.md

@@ -4,10 +4,11 @@
 
   [![Build Status](https://travis-ci.org/G3viz/g3viz.svg?branch=master)](https://travis-ci.org/G3viz/g3viz)
   [![CRAN_version](http://www.r-pkg.org/badges/version/g3viz)](https://cran.r-project.org/package=g3viz)
-  [![CRAN_monthly_download](https://cranlogs.r-pkg.org/badges/g3viz)](https://cran.r-project.org/package=g3viz)
 
-## Check [introduction (live demo)](https://g3viz.github.io/g3viz/)
-## Check [g3viz chart themes (live demo)](https://g3viz.github.io/g3viz/chart_themes.html)
+## Live demo
+1. [Short introduction](https://g3viz.github.io/g3viz/)
+1. [Chart themes](https://g3viz.github.io/g3viz/chart_themes.html)
+1. [An example of showing multiple genes in a shiny app](https://g3viz.github.io/g3viz/shiny_app.html)
 
 ## Introduction
 
@@ -41,3 +42,9 @@ if("devtools" %in% rownames(installed.packages()) == FALSE){
 devtools::install_github("g3viz/g3viz")
 ```
 
+
+## What's new
+
+1.1.3  Update Pfam version 33.1 (dated: 2020-08-24)
+       Update UniProt to 2020-08-24
+       For [Issues #5](https://github.com/G3viz/g3viz/issues/5)

examples/custom_domain_information.R

@@ -0,0 +1,23 @@
+library(g3viz)
+
+#maf.file <- system.file("extdata", "TCGA.BRCA.varscan.somatic.maf.gz", package = "g3viz")
+#mutation.dat <- readMAF(maf.file)
+
+
+domain.fn <- "data-raw/TP53_P04637_interpro.csv"
+domain.df <- read.csv(domain.fn)
+
+Start.col = "Start"
+End.col = "End"
+Domain.ID = "Name"
+
+all(c(Start.col, End.col, Domain.ID) %in% colnames(domain.df))
+
+hgnc2pfam("TP53", output.format = "list")
+
+library(biomaRt)
+
+# select human ensembl
+ensembl = useMart("ensembl",dataset="hsapiens_gene_ensembl")
+
+#