
"To do" list for R/qtl
----------------------------------------------------------------------
This file is intended to contain a list of many of the additions and
revisions that are planned for the R/qtl package.  

If you any additions or revisions to suggest, please send an email to
Karl Broman, <kbroman@biostat.wisc.edu>.
----------------------------------------------------------------------
 
SHORT TERM:

o In c.cross, include an argument for "flipping" a backcross, so that
  a backcross to (AxB)xB can be combined with an intercross

o implement the "controlAcrossCol" argument within summary.scanone,
  to calculate p-values?

o qtl x covar interactions in stepwiseqtl

o scantwo: ability to scan just specific chromosome pairs

o special treatment of X chromosome in scantwo permutations and in
  stepwiseqtl 

----------------------------------------------------------------------

MEDIUM-SHORT TERM:

o get scanone and scantwo to work with a single marker on a
  chromosome. 

o function for getting x axis location in scantwo picture (like the
  new xaxisloc.scanone)

o In mqmsetcofactors(), we might allow cofactors() to be marker names
  rather than just marker indices.

o In the various MQM functions, pheno.col might be a phenotype name
  rather than just numeric index.  Also, it might be a vector of
  phenotype values.  (Both, as in scanone etc.)

o Better way to make use of ... in plot functions that allows
  unmentioned defaults.

o method for having alternative genetic maps and for storing a
  physical map; linear interpolation for plotting scanone output in
  Mbp. 

o make various functions work appropriately with class "special"
  (eg, plot and summary)

o Fix convergence problems in scanone with method="ehk", especially
  in the case of covariates and interactions. 

o Plotting scanone results for *many* phenotypes as an image plot.
  (Perhaps threshold the really high LODs and eliminate curves that
  fail to meet a given threshold.)

o Allow plot.geno (or other cases in which subset.cross is used to
  pull out individuals) to refer to individual IDs.

o Go through all of the various plot functions and make sure that
  the x- and y-axis labels are created with axis() rather than
  text() and segments().  This way, the size of the labels can be
  modified with par(cex.axis)

o Finish off the work to get coefficient estimates by imputation in
  fitqtl for the X chromosome in BC and F2. 

o Revise c.cross so that you can combine crosses even if there are
  different numbers of chromosomes

o 2 traits vs genotype at one QTL (with regression lines)

o Documentation on RI lines.

----------------------------------------------------------------------

MEDIUM TERM:

o eQTL-related stuff:
  
  - Compressing calc.genoprob results (simpler version of locations,
    or just the grid)

  - Mb positions within cross object; multiple genetic maps
  
  - Positions in phenotype object + other annotations

o For the stepwiseqtl function:
   - should be able to save all models with plod within some value
     of the best model 
   - QTL x covariate interactions in countterms
   - it'd be nice to be able to have all models that are one term
     away from the optimal model (maybe this could be a separate
     function) 
   - it'd be nice to be able to easily make a plot of the final model
     plus information about all models that are one term away

o Revise plot.rf so that it can have different color schemes, as in
  plot.scantwo.  Allow a zscale.

o Simpler methods to get at interesting bits in the est.rf results.

o cM coordinates in scantwo plot for multiple chromosomes: Fix
  plot.scantwo for the case that incl.markers=TRUE, so that positions
  are not equally spaced, but are according to the genetic map.  (This
  is working if just one chromosome is plotted, but should be made to
  work generally.)

o scanone (or is it scantwo?) with method="hk": Major memory problems
  in the permutations with multiple phenotypes.

o X chromosome for 4- and 8-way RIL by sibling mating (also 2-way RIL).

o pull all of the genoprob and imputation information out of 
  the qtl objects and replace it with indices to the information
  in the cross object.  (This stuff takes up too much space!)

o lodint/bayesint as option to summary.scanone.

o conditional LOD scores from scantwo output.

o scanqtl: if formula symmetric w/ respect to two QTL that are on the 
  same chromosome, only scan the triangle (rather than the square)

o write tools for converting the output from scanqtl() to the format
  for scanone() or scantwo(), according to whether it's a 1-d or
  2-d search, or print a warning otherwise.

o pairprob at markers + putative QTL?  For linked loci in scanqtl by
  HK/EM. 

o Include Bjarke's code on eHK for scantwo.

o revise read.cross with format="mm" to deal with ril by selfing
  and/or sib mating

o read.cross for "qtx" sometimes doesn't seem to take the
  genotype pattern appropriately; read in a backcross as if it
  were an F2. 

o An NA in the mapmaker data file caused an error in read.cross;
  the line became too long.  Maybe this is true whenever an item
  doesn't match what is expected.

o Speed up read.cross.mm; deliver meaningful errors if map/genotypes
  don't match, and if too many genotypes in a row.

o scanone with additive alleles at QTL

o Pull out results for an interval from scanone.

o Add additional HMM functions for the X chr in RIL, as the marginal
  genotype distribution is 2:1 rather than 1:1 and the transition
  matrix is not symmetric Pr(BB|AA) = 2r/(1+4r) and Pr(AA|BB) =
  4r/(1+4r)

o Add appropriate functions to analyze advanced intercrosses (AILs)
  and advanced backcross (BCn). 

o Allow phenotypes on multiple individuals (esp for recombinant inbred
  lines). 

----------------------------------------------------------------------

LONG TERM:

o covariates with 2part model in scanone

o X chromosome in cim().

o Analysis of censored phenotypes.

o Analysis of residuals.

o Imprinting/parent-of-origin effects.

o Treating a covariate as a random effect.

o Multiple phenotypes (esp. regarding pleiotropy).

o Take the fit of the null model outside of the C code for
  the imputation method in scanone and scantwo, so that it
  only has to be done once (rather than for each chr or chr pair).

o Generalized linear models in scanone and scantwo.

o Incorporate code from Brian Yandell, Fei Zou and Amy Jin on
  semi-parametric QTL mapping methods.

o Analysis functions such as scanone and scantwo might assign an
  attribute to their output which identifies the input data and/or
  function call.

o Re-write the C code for EM underneath scanone and scantwo so that it
  is not so tedious.

----------------------------------------------------------------------
end of TODO.txt
