Human and fly studies tally good and bad mutations, stress ongoing role of natural selection 
John Easton

Researchers from the University of Chicago have demonstrated that 
natural selection plays a much larger role in molecular evolution than 
anyone suspected. Their report, published in the February 28 issue of 
Nature, shows that about 25 percent of genes are evolving rapidly in 
response to competitive pressures. A second paper in the same issue 
confirms this discovery.

Although these papers focus on fruit flies, a previous report from the 
Chicago authors found a similar role for positive and negative selection 
on the human genome. Data from the previous study (Genetics, July 
2001) allowed them to estimate the number of fixed "good" mutations, 
which distinguish humans from monkeys, and the number of residual 
"bad" mutations, genetic flaws that have piled up in the genome and are 
slowly being eliminated. 

These papers directly conflict with the "neutral theory," which has 
dominated genetic research since the 1960s. According to the neutral 
theory, many small genetic changes randomly occur, but the vast 
majority simply don't matter. Fewer than one percent make enough of a 
difference that they are either embraced or expunged by natural selection.

"For several decades, the neutral theory has dominated thinking about 
evolution, but we haven't had the technology to test it," said Chung-I 
Wu, Ph.D., professor and chairman of ecology and evolution the 
University of Chicago and director of the study. "Now we are finding 
that, contrary to this accepted theory, Darwinian forces play a dominant 
role."

To measure the importance of selection at the genetic level, Wu and his 
former graduate students Justin Fay, Ph.D., and Gerald Wyckoff, Ph.D., 
tallied the minute variations within each of 45 genes among flies of one 
species (Drosophila melanogaster) and contrasted them with the same 
genes from a different species (Drosophila simulans).

They found that competitive pressures were shaping about one out of 
four genes. Thirty-four of the 45 genes, or about 75 percent, showed no 
sign of natural selection. But, 11 genes, or about 25 percent showed 
evidence of ongoing rapid evolution. These genes contribute 
disproportionately to the total number of changes, says Wu.

Most of these genes, note the authors, are involved in processes such 
as disease resistance or sexual reproduction, areas where there is 
"continually room for improvement." By studying variation within human 
genes and comparing them with genes from old-world monkeys, Wu's 
team has found that the survival of the fittest is just as active in humans. 

By comparing variation within the human genome and divergence from 
our ape ancestors, the researchers determined that about 35 percent of 
the accumulated changes were "good." "The proportion is shockingly 
high," said Wu, "for someone like myself who grew up in the neutralist 
era." It means one advantageous substitution has entered the human 
genome every two centuries since humans separated from monkeys 30 
million years ago. 

"Humans are getting better," Wu added, "but nobody is perfect." Thirty 
to 40 percent of amino acid changes in human populations, the 
researchers report, are in fact slightly deleterious. They estimate that the 
average human carries about 500 harmful mutations, which are destined 
to be removed from the population by natural selection, but "transiently 
pile up before their exit." The assembly of fruit flies, the Nature paper 
shows, is no less shoddy.

These papers do not mean the end of the neutral theory, cautions Wu. 
But evolutionary geneticists familiar with the work expect these to be the 
first of a long string of papers that will rigorously test the theory and 
determine how much of existing genetic variation matters in the 
competition for survival.

The neutral theory, proposed by geneticist Motoo Kimura in 1968, was 
initially controversial but slowly gained near-gospel status. Before the 
advent of modern genetics, people studied evolution by looking at 
observable differences -- such as variation in the shape of a bird's beak -- 
with a clear fitness value. In the 60s, researchers realized that underneath 
the few obvious differences between related species there were millions 
of DNA variations, far too many for natural selection to sort out. 

So Kimura developed a mathematical framework to explain how evolution 
worked at the genetic level. He argued that the great bulk of DNA 
changes were neutral, biologically insignificant consequences of random 
mutation, and seldom if ever driven by natural selection. "It was a 
simple, elegant, beautiful theory," said Wu, "a nice, clean hypothesis 
that enabled us to make and test predictions based on statistical 
probabilities. But we are now reminded that biology is by nature very 
messy, a historical process that generates variety and accumulates 
multiple tiny aberrations to cope with changing environments." 

"These papers tell us how imperfect our genomes really are," said Wu. 
"At the same time, they tell us how much improvement we have 
constantly been making, all by means of natural selection." 

Oryginal: http://www.uchospitals.edu/news/Evolve.html

------------------------------------------------------------------------

Adaptive protein evolution in Drosophila
NICK G. C. SMITH AND ADAM EYRE-WALKER.
Nature, 415, 1022 - 1024 (28 February 2002) 

For over 30 years a central question in molecular evolution has been 
whether natural selection plays a substantial role in evolution at the 
DNA sequence level. Evidence has accumulated over the last decade 
that adaptive evolution does occur at the protein level, but it has 
remained unclear how prevalent adaptive evolution is. Here we present a 
simple method by which the number of adaptive substitutions can be 
estimated and apply it to data from Drosophila simulans and D. yakuba. 
We estimate that 45% of all amino-acid substitutions have been fixed by 
natural selection, and that on average one adaptive substitution occurs 
every 45 years in these species.
------------------------------------------------------------------------

Testing the neutral theory of molecular evolution with genomic data from Drosophila
JUSTIN C. FAY, GERALD J. WYCKOFF & CHUNG-I WU.
Nature, 415, 1024 - 1026 (28 February 2002) 

Although positive selection has been detected in many genes, its overall 
contribution to protein evolution is debatable. If the bulk of molecular 
evolution is neutral, then the ratio of amino-acid (A) to synonymous (S) 
polymorphism should, on average, equal that of divergence. A 
comparison of the A/S ratio of polymorphism in Drosophila 
melanogaster with that of divergence from Drosophila simulans shows 
that the A/S ratio of divergence is twice as high a difference that is 
often attributed to positive selection. But an increase in selective 
constraint owing to an increase in effective population size could also 
explain this observation, and, if so, all genes should be affected similarly. 
Here we show that the difference between polymorphism and divergence 
is limited to only a fraction of the genes, which are also evolving more 
rapidly, and this implies that positive selection is responsible. A higher 
A/S ratio of divergence than of polymorphism is also observed in other 
species, which suggests a rate of adaptive evolution that is far higher 
than permitted by the neutral theory of molecular evolution.

POWRÓT