R/qtlbim: Quantitative Trait Loci with Bayesian Interval Mapping

Please redirect to the official www.qtlbim.org page for the most up-to-date information on R/qtlbim. We are developing a Frequently Asked Questions page. The CRAN contributed package page is http://cran.us.r-project.org/web/packages/qtlbim/

NOTE: This pages serves for three related pieces of software. The first two rely on R/qtl (Broman et al. 2003 Bioinformatic), also available from CRAN. It is assumed that you know how to create a cross object using R/qtl.

• R/qtlbim was released via CRAN in September 2006 [be sure to get version 1.05-6 or later]. This project involved a collaboration of Nengjun Yi (U AL Birmingham) and Brian Yandell (U WI Madison), along with Samprit Banerjee, Daniel Shriner, and Tapan Mehta (UAB), Whipple Neely (UW-Madison) and Randy von Smith (Jackson Labs). This is the preferred software for new and continuing users. It allows for epistasis and interacting covariates, and is considerably faster than earlier code.

  We are developing a Frequently Asked Questions page. For those eager to get going, look at the vignettes that come with the package. And you can try the following:

  library(qtlbim)
  demo(qb.tour)
  

  To run your cross object without further ado:

  > library(qtlbim)
  ## Set up genotype probabilities at pseudomarkers.
  > cross <- qb.genoprob(cross, step = 2)
  ## Create MCMC samples within R.
  > myqb <- qb.mcmc(cross, pheno.col = 1)
  ## Summary and plot.
  > summary(myqb)
  > plot(myqb)
  

• R/bim has been in CRAN and Bioconductor for several years. It was developed by Brian Yandell, Jaya Satagopan (Sloan-Kettering Mem Inst) and Pat Gaffney (Imclone), with important contributions from Hao Wu (Jackson Labs, now a student at Johns Hopkins U) and input from Chunfang "Amy" Jin (McKinsey & Co.).

  > library(bim)
  ## Create MCMC samples within R.
  > mybim <- run.bmapqtl(cross, pheno = 1)
  ## Summary and plot.
  > summary(mybim)
  > plot(mybim)
  

• Bmapqtl module of QTLCartographer. The output of the Bmapqtl can be the input to R/bim via the read.bim() routine. Currently that is a bit of a kludge, requiring a file called nval.dat. You may have to do the following:

  > library(bim)
  ## Initialize options.
  > bmapqtl.options()
  ## Write options to file "nval.dat". You can modify values using
  ## write.bmapqtl() or later edit the file by hand.
  > write.bmapqtl()
  ## Read your Bmapqtl file created from QTLCart.
  > mybim <- read.bim(".", "yourbmapqtlfilename")
  ## Summary and plot.
  > summary(mybim)
  > plot(mybim)
  

References

• Gaffney PJ (2001) An efficient reversible jump Markov chain Monte Carlo approach to detect multiple loci and their effects in inbred crosses. PhD thesis, UW-Madison Statistics. [Paper]
• Satagopan JM, Yandell BS (1996) Estimating the number of quantitative trait loci via Bayesian model determination. Special Contributed Paper Session on Genetic Analysis of Quantitative Traits and Complex Diseases, Biometric Section, Joint Statistical Meetings, Chicago, IL. [Paper]
• Satagopan JM, Yandell BS, Newton MA, Osborn TC (1996) Markov chain Monte Carlo approach to detect polygene loci for complex traits. Genetics 144: 805-816. [Paper]
• Yi N, Xu S, George V, Allison DB (2004) Mapping multiple quantitative trait loci for ordinal traits. Behavior Genetics 34: 3-15. [Paper]
• Yi N, Yandell BS, Churchill GA, Allison DB, Eisen EJ, Pomp D (2005) Bayesian model selection for genome-wide epistatic quantitative trait loci analysis. Genetics 170: 1333-1344. [Paper]