Page 244 - Applied Probability

11
Radiation Hybrid Mapping
11.1 Introduction

In the 1970s Goss and Harris [12] developed a new method for mapping
human chromosomes. This method was based on irradiating human cells,
rescuing some of the irradiated cells by hybridization to rodent cells, and
analyzing the hybrid cells for surviving fragments of a particular human
chromosome. For various technical reasons, radiation hybrid mapping
languished for nearly a decade and a half until revived by Cox et al. [10].
The current, more sophisticated and successful versions raise many fasci-
nating statistical problems. We will first discuss the mathematically simpler
case of haploid radiation hybrids. Once this case is thoroughly digested, we
will turn to the mathematically subtler case of polyploid radiation hybrids.
In the haploid version of radiation hybrid mapping, an experiment starts
with a human–rodent hybrid cell line [10]. This cell line incorporates a full
rodent genome and a single copy of one of the human chromosomes. To frag-
ment the human chromosome, the cell line is subjected to an intense dose of
X-rays, which naturally also fragments the rodent chromosomes. The repair
mechanisms of the cells rapidly heal chromosome breaks, and the human
chromosome fragments are typically translocated or inserted into rodent
chromosomes. However, the damage done by irradiation is lethal to the cell
line unless further action is taken to rescue individual cells. The remedy is
to fuse the irradiated cells with cells from a second unirradiated rodent cell
line. The second cell line contains only rodent chromosomes, so no confu-
sion about the source of the human chromosome fragments can arise for a
new hybrid cell created by the fusion of two cells from the two different cell
lines. The new hybrid cells have no particular growth advantage over the
more numerous unfused cells of the second cell line. However, if cells from
the second cell line lack an enzyme such as hypoxanthine phosphoribosyl
transferase (HPRT) or thymidine kinase (TK), both the unfused and the
hybrid cells can be grown in a selective medium that kills the unfused cells
[10]. This selection process leaves a few hybrid cells, and each of the hybrid
cells serves as a progenitor of a clone of identical cells.
Each clone can be assayed for the presence or absence of various human
markers on the original human chromosome. Depending on the radiation
dose and other experimental conditions, the cells of a clone generally con-
tain from 20 to 60 percent of the human chromosome fragments generated
by the irradiation of its ancestral human–rodent hybrid cell [8, 10]. The
basic premise of radiation hybrid mapping is that the closer two loci are