title	author	date	output
Process Sanger data for CC Founders	Brian S Yandell	8 December 2016	html_document

These files are getting features for use with DOQTL data. These are derived from code developed by Karl Broman.

• R/get_mgi_features.R: get MGI features (genes, exons, ...) for Sanger file
• inst/CreatSQL/vcf_snp_2db.R: (re)create cc_foundersnps.sqlite (update of Karl Broman's R/0_vcf2db.R)
• inst/CreatSQL/vcf_indel_2db.R: create cc_founderindels.sqlite (new)
• inst/CreatSQL/svs.Rmd: create svs8_*.rds files (new)

The vcf and svs files have hard-wired dirpath that needs to be locally edited.

Required packages

For MGI features:

library(dplyr)
library(readr)
source("check_interval.R")

For VCF:

library(VariantAnnotation)
library(RSQLite)

For SVS:

library(dplyr)

SQLite Files

The MGI features are in their own SQLite. This is much smaller. Not clear that it is needed this way.

I modified Karl's SNP SQLite code to include consequence and Ensembl IDs. The InDel SQLite code is a minor change to get indels. Note that there is a column labelled allele, which is the name used in creation of the VCF, but this is pretty close to the column alleles, which we use in DOQTL stuff.

SVS Files

The structural variants were small, but unclear what users might want. I mainly use the svs8_len.rds file. These are in a different format from VCF. Look carefully at this. Not totally happy, but it works.

Installation

You can install R/CCSanger from GitHub.

You will need the following packages for CCSanger:

install.packages(c("assertthat", "dplyr", "feather", "dbplyr"))

Once you have installed these, install CCSanger as

install_github("byandell/CCSanger")

---