Monday, February 22, 2016

The winners write the history!

See text in The Scientist on a criticism of Mendelian genetics by Walter Frank Raphael Weldon:

Weldon also took a hard statistical look at Mendel’s data, and the analysis only added to his suspicion. Even if Weldon assumed Mendel’s predictions were accurate, the data were just too good to be true—the equivalent of flipping a coin 100 times and getting exactly 50 heads and 50 tails.

In 1902, Weldon published his critiques in Biometrika, a journal he had cofounded the previous year. According to Radick, Weldon considered Mendel’s data symptomatic of the wider problem with binary categories, which construct a misleading concept of genetic dominance. “Weldon’s confirmed view was that biological science was just on the verge of breaking through to the next level because it was taking variation so seriously,” says Radick. “But it all depended on researchers who were not satisfied with a fictional idealization of complex data—which is, in his mind, what you’re dealing with when you’re dealing with ‘yellow.’ ”

For Weldon, Mendel’s patterns represented special cases where humans had artificially bred out variation in domestic plants and animals. But his paper was met with harsh criticism from a colleague, William Bateson. Former schoolmates, Weldon and Bateson suffered a falling out in the mid-1890s after a series of scientific disagreements. Their rivalry continued through the early 1900s, when Weldon started to write a book detailing his thoughts on Mendel—but Weldon died unexpectedly in 1906 at the age of 46. Bateson continued to champion Mendelian principles and garnered widespread acceptance of rules still taught today in the opening chapters of genetics textbooks.

“Weldon gets remembered, if at all, as the one who misguidedly obstructed the path to Mendelism,” says Radick. “The winners write the history.” But Weldon’s ideas and his unfinished book, stored in the archives at University College London, suggest an alternative path that the early study of genetics might have taken—one that would have more closely approached a modern understanding of DNA’s complexity, embracing variation and unexpected results as crucial to the system rather than exceptions.


Wikipedia notes:

As a project for a History of Statistics course taught at the University of Chicago, Zacariah Labby[6] built a machine[7] to roll the dice and automatically count the dots (pips) on each die. The resulting data allowed Labby to repeat Weldon and Pearson's original investigations, as well as delve deeper into the analysis.

Recall the earlier finding:

The conclusion is that the dice show
a clear bias toward fives and sixes, which
Pearson estimated was probably due to
the construction of the dice. Most inexpensive
dice have hollowed-out pips,
and since opposite sides add to seven,
the face with six pips is lighter than its
opposing face, which has only one pip.


Post a Comment

<< Home