Val Dusek
In a message dated 11/16/99 7:10:58 AM Eastern Standard Time, Colin McLarty
<cxm7@po.cwru.edu> writes:
<< The terrific thing for genetics research was that linkage is not absolute.
Linked traits are sometimes inherited separately, and the frequency of
separation was quite lawlike.
For example, geneticists might find three linked traits, say ABC, such
that sometimes AC was inherited without B, and sometimes CB without A, but
only very rarely AC without B. And in this case there would be rather
definite proportions: CB is inherited say 3.27 times as often as AC. That
suggested to researchers that the traits occur on one chromosome with C in
between A and B, with the AC interval 3.27 times as long as the CB
interval--and thus 3.27 times as likely to be separated.
In fact researchers had many genes in each linkage family and so a great
many different interval-ratios to measure, and with Drosophila they had a
huge amount of data. The many interval ratios made a very good fit to
linear congruence--for example, if another gene D occurred in the order
ADCB then any two of the intervals AD and AC and AB and DC and DB and CB
could be separately compared and the results dependably came out that
interval AC was the sum of AD and DC to high degree of accuracy, DB the sum
of DC and CB, and so on >>