Haplotype exclusion by gene fusion errors based on
the Mouse system
The principle: The program Hapex, for haplotype
exclusion,uses a cellular automaton approach to calculate the gene rearrangement
events in a large sample of cells. The less likely a gene rearrangement
pattern, the more cells have to be looked at. At a rough estimate there
is little of likely interest when cell numbers exceed one million.
The idea is to pick your favorite model and generate
the rearrangement patterns that result, then see if the patterns fit your
The STOP condition is either H-stop-H (and LH-stop
at the light chain loci) or LH-stop for all loci.
The Leakiness of the Stop signal. A value of 0.0
is no leak and a value of 1.0 is stopless.
The number of loci to be analyzed ranges from 2 (H
and Kappa) to 4 (H, Kappa, Lambda 1, and Lambda 2).
The branching ratio that describes the relative probability
of a gene rearrangement event being initiated at a given locus. For example,
a Heavy to Kappa branching ratio of 3 would mean that gene fusions are
initiated at the H locus three time more often than at kappa. A branching
ratio of more than 100:1 is the functional equivalent of a sequential ordering
of the rearrangements.
The choice of gene fusion efficiencies requires some
adjustments that include the effect of pseudo V segments (for example if
30% of V segments are defective but fusable, then the fusion efficiency
of 1/3 drops to around 2/9). If one wishes to take into account effects
of receptor editing that allow some productive fusions to be made to reinitiate,
this is done by recalculating the effect on the gene fusion efficiency.
If half of the productive rearrangements were randomly considered to be
non-functional as rearrangements reinitiated, this would result in the
fusion efficiency that is typed in to the program to be 50% lower to reflect
the 50% loss of productive joints.
There are up to 9 sets of gene rearrangement configurations
that you can monitor. The exact choice is offered on a new page.
Checking the box for fine structure analysis of J-Kappa
allows various kinds of gene fusion events to be monitored at the level
of J-Kappa usage.
There are two options for the number of gene fusion
events at J-Kappa, one - meaning only one rearrangement is allowed at the
locus, or many - meaning that the J-Kappas are used until there are none
left as would occur when J-Kappa-4 is rearranged. The J-Kappa loci are
numbered 1-4 based on 1 being the locus immediately 5' to the V-kappa genes
and 4 being the locus immediately 3' to the C-Kappa locus.
If many rearrangements are allowed, they can be sequention
from J-1 to J-4 or they can be random.
The number of cells to be simulated should be around
10^4 at first, then for rare events around 10^6 should be used, though
the time taken to simulate a million sets of gene fusion events can be
quite long and thus we advise looking at smaller samples to get the rough
After clicking the submit button the option list
is set out as a series of check boxes that are self-explanatory.
If your model is not accommodated or you have
results that seem to be unsimulatable with the options provided, please
let us know so that we can adjust the program to simulate your model or
the data. Eventually we hope that only one model will simulate all of the
Return to CIG Home Page