Rqtl2 vs Rqtl

197 views
Skip to first unread message

Othmane Lamoumni

unread,
Oct 7, 2023, 3:13:53 PM10/7/23
to R/qtl discussion
Hello everyone,

I'm new to using Rqtl2 for initial QTL mapping analysis (Interval Mapping) and I'm looking for guidance on how to perform MQM. While I found multiple functions for MQM in Rqtl, my main reason for using an R package for QTL mapping is to overcome computational limits of older Windows Interface software. Rqtl2 offers the 'cores =' argument in key functions, which is crucial for handling larger datasets efficiently. But on the other hand I'm still not able to find MQM functions, or how to conduct it in its tutorials.

As a beginner, I came here to seek help:

  1. Which package version should I use? I believe Rqtl2 is the right choice since I'm working with high-density data.

  2. How can I perform MQM using Rqtl2?

Thank you for your assistance.

Best regards,"


Othmane

Karl Broman

unread,
Oct 7, 2023, 4:15:32 PM10/7/23
to R/qtl discussion
R/qtl2 handle more types of crosses. If you are working with multi-parent populations like MAGIC lines, Collaborative Cross, heterogeneous stock, diversity outbred mice, you will need to use R/qtl2.
But if you are focusing on a traditional two-parent cross such as a backcross or intercross, either R/qtl1 or R/qtl2 may work for you.

R/qtl2 includes fit of linear mixed models to account for population structure. So if you want to use a LMM to account for varying relationships, you'll need to use R/qtl2.

But R/qtl1 has many multiple-QTL model methods, including the MQM approach you mentioned, and these have not yet been implemented in R/qtl2. So if you're using a two-parent cross and want to use MQM, you should stick with R/qtl1.

karl

Othmane Lamoumni

unread,
Oct 8, 2023, 6:49:22 AM10/8/23
to rqtl...@googlegroups.com
Thank you for your clarification
I really appreciate the help 

Othmane 

--
You received this message because you are subscribed to a topic in the Google Groups "R/qtl discussion" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/rqtl-disc/Ii0ZtqdC0w0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to rqtl-disc+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rqtl-disc/0dc090b9-972f-41f7-b8cf-903cb0b29571n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages