From jillian.maddoxalumni.unimelb.edu.au  Mon Jan  7 19:37:28 2013
From: Jill Maddox <jillian.maddoxalumni.unimelb.edu.au>
To: Multiple Recipients of <crimap-usersanimalgenome.org>
Subject: CRI-MAP twopoint bugs - 3-generation, full-sibling families
Date: Mon, 07 Jan 2013 19:37:28 -0600

Hi All

We fixed a twopoint bug in 2.504 but unfortunately it turns out that we
only fixed part of the twopoint bug and that the fix we used made another
part of the bug worse. The bug that we are aware of occurs with 3-
generation full-sibling families with genotyped grandparents, same
heterozygote parents (for each locus the genotype for each parent is 1 2) ,
and occurs with certain phase situations - one parent has allele phasing
based on grandparent genotypes like 2 2 maternal
--- 
1 1 paternal 

and the other parent has allele phasing like 1 2 maternal
--- 
2 1 paternal 

In versions of CRI-MAP 2.503 and earlier the correct theta is found by
twopoint for the remaining bug but for some phase situations the lods are
much lower than they should be. This means that some markers that should be
seen as linked aren't and is mainly likely to be a problem when building
chromosome groups. The markers most likely to be affected will be markers
at the ends of groups as there are more chances to have phase situations
that are OK when more than one marker links something into a group or there
are multiple links between groups of markers.

In version 2.504/2.504a the consequence of the bug is that theta is smaller
than it should be and the lod (which is based on theta) is thus higher than
it should be. This results in false linkage between some markers.

The bug is less of a problem for multi-point situations as the phasing for
ambiguous situations depends more on neighbours' phases.

We are working to fix this bug properly but are not sure when we will have
a fix.

If anyone is aware of a source of funding for further work to improve CRI-
MAP please let us know. My work on CRI-MAP has been funded by a USDA grant
but the funding finished 31 Dec 2012, Ian has volunteered his time.

Regards

 
Jill 

 
*************************************************************** 

Jill Maddox 16 Park Square Port Melbourne, 3207 Australia phone: 03 9646
0428 E-mail: jillian.maddoxalumni.unimelb.edu.au

***************************************************************