MDD OpenGWAS upload

mdd_upload.r

library(GwasDataImport)
library(data.table)
library(ieugwasr)
library(dplyr)


# Download the Howard et al 2019 MDD GWAS from https://datashare.ed.ac.uk/handle/10283/3203
a <- data.table::fread("https://datashare.ed.ac.uk/bitstream/handle/10283/3203/PGC_UKB_depression_genome-wide.txt?sequence=3&isAllowed=y")

# It doesn't have chr/pos, so let's look them up on OpenGWAS (a bit slow!)
b <- get_positions(a$MarkerName)

# Merge in the chr/pos
ab <- dplyr::inner_join(a, b, by=c("MarkerName"="rsid"))

# Write the file
data.table::fwrite(ab, "PGC_UKB_depression_genome-wide_chrpos.txt.gz", sep=" ", quote=FALSE, row=FALSE, col=TRUE)

# Initialise
x <- Dataset$new(filename="PGC_UKB_depression_genome-wide_chrpos.txt.gz")

# Specify columns
x$determine_columns(list(chr_col=8, snp_col=1, pos_col=9, oa_col=3, ea_col=2, eaf_col=4, beta_col=5, se_col=6, pval_col=7))

# Process dataset
x$format_dataset()

# Input metadata
x$collect_metadata(list(
	trait="Major depression",
	author="Howard DM",
	consortium="PGC",
	year="2019",
	pmid="30718901",
	build="HG19/GRCh37",
	category="Disease",
	subcategory="NA",
	group_name="public",
	population="European",
	sex="Males and Females",
	note='UKBiobank and PGC (excluding 23andme)',
	ontology="EFO:0003761;MONDO:0002009",
	ncase=170756,
	ncontrol=329443,
	unit="logOR"))

# Upload metadata
x$api_metadata_upload()

# Upload summary data
x$api_gwasdata_upload()

# View report
x$api_report()

# Release dataset
x$api_gwas_release()

# Delete wd
x$delete_wd()

x <- Dataset$new(igd_id="ieu-b-102")
x$api_report()
x$collect_metadata(list(
	trait="Major depression",
	author="Howard DM",
	consortium="PGC",
	year="2019",
	pmid="30718901",
	build="HG19/GRCh37",
	category="Disease",
	subcategory="NA",
	group_name="public",
	population="European",
	sex="Males and Females",
	note='UKBiobank and PGC (excluding 23andme)',
	ontology="EFO:0003761;MONDO:0002009",
	ncase=170756,
	ncontrol=329443,
	unit="logOR"))

x$api_metadata_edit()