QQ Plot recipe for PLCO

Start here with your VPN on: https://exploregwas-dev.cancer.gov/plco-atlas . Note APIs.

"For the QQ plots, we perform two queries. The first query retrieves the top 10,000 points sorted by lowest p-value (using the /variants route). The second query (using the /points route) returns a subset of points which is sufficient to draw the remainder of the plot (eg: it will look visually identical to a plot which includes all variants).

The Manhattan plots use two routes: for the summary view (all chromosomes), we use the /summary route, passing in the phenotype_id, sex, and minimum -log(p) value (nlogp). For the exploregwas site, this value is set to 2.

For the chromosome-level view, we use the /variants route, also passing in the phenotype_id, sex, and nlogp. Optionally, we can include additional parameters to further customize the response from the api (for a more in-depth overview, you can view the api documentation at: https://exploregwas-dev.cancer.gov/plco-atlas/#/api-access). For returning large datasets, it is recommended to set raw=true, as that greatly reduces the time it takes to receive a response."

Let's give it a try starting with raw VCF source

Confluence

... Tom?

Ayo

csv conversions

excell original : https://nih.box.com/s/0d2t85428dl815ulz89a5ffkpvkxyhe0

csv: https://nih.box.com/s/1qwopu4w2cnxn46h3x91jgi72k4akq4s