Skip to content

Latest commit

 

History

History
136 lines (95 loc) · 4.45 KB

README.md

File metadata and controls

136 lines (95 loc) · 4.45 KB
Raw

mmconvert

R-CMD-check CRAN_Status_Badge zenodo DOI

Karl Broman

R package to convert mouse genome positions between build 39 physical locations and the Cox genetic map positions. (See the Cox map version 3, updated for the build 39 physical map.)

The package is a reimplementation of part of the basic functionality of the mouse map converter web service from Gary Churchill's group at the Jackson Lab.

Installation

Install the package from CRAN with install.packages("mmconvert").

Alternatively, install the mmconvert package from GitHub using the remotes package:

install.packages("remotes")
remotes::install_github("rqtl/mmconvert")

Usage

mmconvert contains two functions: mmconvert() and cross2_to_grcm39().

mmconvert()

mmconvert() takes a set of positions as input, plus an indication of whether they are basepairs or Mbp (in build 39) or sex-averaged, female, or male cM from a slightly smoothed version of the revised Cox genetic map.

The input positions can be character strings like "chr:position".

input_char <- c(rs13482072="14:6738536", rs13482231="14:67215850", gnf14.117.278="14:121955310")

Or they can be a list of marker positions, separated by chromosome.

input_list <- list("14"=c(rs13482072=6738536, rs13482231=67215850, gnf14.117.278=121955310))

Or they can be a data frame with chromosome IDs as the first column and positions as the second column. Marker names can either be in a third column or included as row names.

input_df <- data.frame(chr=c(14,14,14),
               pos=c(6738536, 67215850, 121955310),
               marker=c("rs13482072", "rs13482231", "gnf14.117.278"))

For any of these cases, the output is a data frame with seven columns: marker, chromosome, sex-averaged cM, female cM, male cM, basepairs, and mega-basepairs.

library(mmconvert)
mmconvert(input_df)
##                      marker chr cM_coxV3_ave cM_coxV3_female cM_coxV3_male bp_grcm39 Mbp_grcm39
## rs13482072       rs13482072  14    0.3938878       0.4204617     0.3707116   6738536   6.738536
## rs13482231       rs13482231  14   28.7007118      34.8398609    22.8635376  67215850  67.215850
## gnf14.117.278 gnf14.117.278  14   59.5630160      64.5897950    55.0750080 121955310 121.955310

If you want to give the input positions in Mbp rather than basepairs, use the argument input_type="Mbp".

input_df$pos <- input_df$pos / 1e6
mmconvert(input_df, input_type="Mbp")
##                      marker chr cM_coxV3_ave cM_coxV3_female cM_coxV3_male bp_grcm39 Mbp_grcm39
## rs13482072       rs13482072  14    0.3938878       0.4204617     0.3707116   6738536   6.738536
## rs13482231       rs13482231  14   28.7007118      34.8398609    22.8635376  67215850  67.215850
## gnf14.117.278 gnf14.117.278  14   59.5630160      64.5897950    55.0750080 121955310 121.955310

The input positions can also be provided in sex-averaged, female, or male cM. But note that the bp or Mbp positions must be in mouse genome build 39, and cM positions must be according to the Cox Map V3.

cross2_to_grcm39()

cross2_to_grcm39() takes a cross2 object from R/qtl2 with mouse genotype data from one of the MUGA arrays and converts it to mouse genome build GRCm39, by possibly subsetting the markers, reordering them according to the GRCm39 build, and plugging in GRCm39 Mbp positions and the revised Cox genetic map. See https://github.com/kbroman/MUGAarrays for the MUGA array annotations and https://github.com/kbroman/CoxMapV3 for the revised Cox genetic map.

file <- paste0("https://raw.githubusercontent.com/rqtl/",
               "qtl2data/main/DOex/DOex.zip")

library(qtl2)
DOex <- read_cross2(file)

DOex_rev <- cross2_to_grcm39(DOex)

License

Licensed under GPL-3.