Biology Reference
In-Depth Information
(for example, men are more likely to have heart attacks than
women), whereas 23andMe primarily takes into account age
(for example, incidence of rheumatoid arthritis increases with
age). This ambiguity in the defi nition of a “population” under-
scores the caution one must exercise when interpreting absolute
risk results.
49
Although Navigenics and 23andMe are reporting risks, the crucial
question is how such risks should be interpreted. Their interpretation
relies on statistics—probabilities, averages, sampling, and so on. Un-
derstanding the body as a risky body entails fi rst understanding it as
a statistical one. Whereas the concept of “genetic risk” points toward
genetic fatalism, the statistical body suggests something more open-
ended—especially the possibility of using personal genomics to take
control of one's own biological destiny. Moreover, the statistical body
is not an individual body—based on a set of individualized risks—but
a body that exists only in relation to other bodies (and in relation to
a population). The statistical body carefully crafts “normalcy” (from
intricately contrived mathematical and statistical analyses of genomes)
and locates particular bodies in relation to that ideal. The original “hu-
man genome,” reconstructed from the DNA of a few individuals and
based on no data on human genetic variation, had very little power to
claim that it represented an “ideal” or “normal” genome. With personal
genomics, the 1000 Genomes Project, and other similar endeavors, the
“normal” emerges clearly and powerfully from mathematical averages.
Each one of us may soon be able to defi ne ourselves statistically in rela-
tion to these means and to live out our lives accordingly.
Computers and networks have not only made this sort of work pos-
sible on a practical level, but have also generated a particular set of
tools and practices and sensibilities for working with biological data.
In personal genomics, we are just beginning to see how these tools and
practices and sensibilities—which I have called bioinformatic—are
manifesting themselves in health care. Problems of statistics, of manag-
ing massive amounts of data, of computing probabilities based on these
data, are doing much more than opening up possibilities for preventa-
tive medicine. In the study of variation, the statistical body may come
to describe a particular and powerful “normality” that is derived from
probabilistic models that discover patterns among the vast amounts of
disordered data from genomics and medical informatics. That ideal is
likely to transform our own sociality—how we understand the simi-
larities and differences between us and those around us—and our in-
