R/qtl discussion Google Group

Running CIM with multiple phenotypes

2013-05-17T01:56:53Z

I have a quick question. Is it possible to implement mapping on
multiple traits at once with composite interval mapping? For scanone,
I can run multiple phenotypes using pheno.col, such as
pheno.col=1:5000. This is computationally efficient, especially when
parallelizing across multiple processors using snow. However, I cannot

write.cross() for BCsFt crosses

2013-04-29T19:16:17Z

Hi all,
I would like to use the write.cross() function for a BCsFt type cross, but it only seems to work for "gary" and "qtlcart" formats. Is this pilot error on my part, or are the other outputs not supported for BCsFt type crosses?
If it's not supported, does anyone have a bit of code that can combine and format the outputs of pull.geno(), pull.pheno(), and pull.map() into the csv/csvr format?

Using rqtl with GWAS data

2013-04-24T13:46:38Z

Dear Karl,
I work with GWAS data in mouse and I've found some very interesting options
in rqtl that I would like to explore with my data.
I have a very dense marker coverage of my genome (~200.000 snps, ~1snp
every 4Kb), and therefore I dont need to perform the snp imputation step. I
took a look into the manual and it specifies that all analysis require the

multiple testing correction for multiple trait

2013-04-22T19:55:24Z

Dear Karl,
I am currently trying to map QTL for multiple traits treating them as each
trait separately as a phenotype. Permutation scheme is applied for each
trait individually and the significance threshold is calculated. The genome
wide significance defines the threshold for significant QTL for each

coding 3-way cross as 4-way cross?

2013-04-12T00:38:56Z

Hi Karl,
I have a set of ~1200 markers that are AA/CD in the female/male parents
respectively, and I also have the genotype data for 16 F1 individuals.
Using this data, I would like to see if there is a difference in
recombination frequency between males and females. To do this though, I was
first planning to create a linkage map for these markers so that I can see

Re: R 3.0 compatibility issues with R/qtl

2013-04-11T19:12:52Z

I'm not exactly sure what's going on and what you should do.
Maybe try a different repository?

options(repos="[link]")
install.packages("qtl")

karl

R/qtl version 1.27-10

2013-04-11T18:38:39Z

A new version of R/qtl (1.27-10) has been released at its web site
([link]) and will be on CRAN shortly.
The main change is the implementation (by Laura Shannon and Brian
Yandell) of tools for advanced backcrosses/intercrosses. See the
tutorial at [link]

R 3.0

2013-04-11T02:15:59Z

Hi R/QTL team,
Just to warn some of you who might be interested in the new version of R
(3.0) recently released, qtl library is not compatible yet. So better stick
to v2.15
Jo

not enough columns in excel to create the input file

2013-04-05T20:50:33Z

This is my second time using R/qtl. The first time I had a very small cross but now I have 300,000 markers. I realize I cannot set up the input file like I did last time because there are not enough columns in excel. Does anyone have advice for I can get enough columns to create my input file? Can I generate the file transposed (there are enough rows in excel to do this)? Thanks and sorry for the basic question!

Scan one problem with non-numeric phenotype error

2013-04-04T18:20:29Z

Dear Karl and group

I am trying to run scanone on a dataset with 3 phenotypes, and I am
getting an error:

Error in checkcovar(cross, pheno.col, addcovar, intcovar, perm.strata,

Following phenotypes are not numeric: Column 1

I get the same error when I specify any of the "real" phenotype columns

Two QTL scans

2013-04-03T04:07:23Z

Dear Karl and the Group,

I am a little confused one how the thresholds for the 'two QTL scan' were
determined as 6.0, 4.7, 4.4, 4.7, 2.6 for mouse backcross (using the hyper
data) at 5% significance for *Tf*, *Ta*, *Ti*, *Tfv1*, *Tav1*.

I am unclear what is meant by the following text in your book (in Bold);

Definition of QTL

2013-03-31T17:45:20Z

Hi Kar,
I am using R/qtl to call QTLs but I get confused about the definition of
QTL.
Just consider single QTL model:
Can one QTL contain several markers whose genotype may differ in several
individuals? Or QTL can only be defined as a marker that is nearest to the
peak of LOD?
Thanks!
Best
Yupeng

Scan QTL for RIL data with quarter AB

2013-03-29T18:48:46Z

Hi all,
I am working on data of F4 RILs by selfing and looking for QTLs from this phenotype-genotype data. Since the generation of these RILs is small, there are lots of markers with value "heterozygous". In the data, the composition of genotype is AA:36.4 AB:26.7 BB:36.9
Since the data is from RILs, I tried convert2riself() and it discarded all the ABs. I am concerning that throwing away ~26% genotypes may dramatically reduce the statistical power to detect QTLs. For the low-generation RIL data, is it fair to treat it as intercross?

imputed genotypes

2013-03-13T12:49:10Z

Hi, i am making QTL analysis with both codominant markers (microsatellite)
and dominant markers (AFLPs) and when I draw the effect plots for each QTL
I get the genotypes that were imputed plotted in red. I would be interested
to know who those red genotypes are, i.e. I would like to be able to plot

Changing population type

2013-03-12T20:57:40Z

Hi there,
I'm pretty brand new to Rqtl and I am trying to create a linkage map for an IBL (inbred backcross population) of 75 individuals. When R reads my map, I get the following output:
--Read the following data:
75 individuals
630 markers
1 phenotypes
--Estimating genetic map