Using ATGC genotype format

77 views
Skip to first unread message

Akinlade Oluwaseun Johnson

unread,
Oct 19, 2021, 7:28:36 AM10/19/21
to R/qtl2 discussion
Hi,

I am trying to use QTL2 for a MAGIC population developed from 8 founders but my genotype file is in ATGC format. Is it possible to use this format? Secondly, I do not have the genetic map I only have physical map, is it possible to run the analysis with just the physical map?

Thank you and best regards,
Olujohnson 

Karl Broman

unread,
Oct 19, 2021, 7:31:36 AM10/19/21
to R/qtl2 discussion
Yes, it needs to be converted from ATGC to like A/H/B or 1/2/3.

For qtl analysis, you ultimately need an estimate of a genetic map, but you could start by using the physical map in Mbp, or use some estimate of the recombination rate to convert Mbp to cM.

karl


--
You received this message because you are subscribed to the Google Groups "R/qtl2 discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rqtl2-disc+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rqtl2-disc/fb53ef5b-c151-4e7a-9a84-5394edbd42f4n%40googlegroups.com.

Olujohnson

unread,
Oct 19, 2021, 5:07:25 PM10/19/21
to R/qtl2 discussion
Thank you so much, Karl.

I will give it a try and update the group on my progress.

Please is there any function in qtl2 to estimate the recombination rate?

Thank you and best regards,
olujohnson



Karl Broman

unread,
Oct 19, 2021, 5:19:34 PM10/19/21
to R/qtl2 discussion
You can use the function est_map() to estimate the inter-marker distances.

Here's an example for the 19-way Arabidopsis MAGIC lines.

library(qtl2)
file <- paste0("https://raw.githubusercontent.com/rqtl/",
               "qtl2data/main/ArabMAGIC/arabmagic_tair9.zip")
arab <- read_cross2(file)
map <- est_map(arab, quiet=FALSE) # can use cores=0 to use multiple CPU cores
arab$gmap <- map

It may take a super long time.

karl

Olujohnson

unread,
Oct 25, 2021, 6:47:51 AM10/25/21
to R/qtl2 discussion
Thank you so much, Karl.

Reply all
Reply to author
Forward
0 new messages