据physorg网站2007年10月10日报道，美国威斯康辛大学麦迪逊分校研究人员发现了分子所具有的一项绝技，他们在基因层面上详细阐述了自然选择的过程。
霍华德.休医学研究所研究员肖恩.B.卡罗尔和前美国威斯康辛大学麦迪逊分校研究生克里斯.托德.黑亭基尔在《自然》杂志上撰写了一篇论文，详细阐述了单个酵母基因是如果在许多代时间内分裂成两个，以便更加有效地适应环境。他们的研究显示，基因分裂是推动进化的最基本动力。卡罗尔是全球著名的进化生物学家，他说，“这是新能力出现，新功能进化发展的原因。蝴蝶、大象和人类身上正在发生这种变化。进化一直在进行之中。”
此项研究意义重大，是因为它让我们从最基层面了解了生物体变得更加适应他们所处环境的方式。研究为我们揭示了自然选择的工作方式。自然选择这一重要的理论最先由查尔斯.达尔文提出，达尔文认为积聚无规则变化，然后量变引发质变，从而提升生物体的生存能力，被自然“选择”遗传给未来后代。
此项新研究重新演示了1亿年前或几年前酵母系列基因的改变，当一个关键基因被复制时，它会将它的营养处理反应分开来，以便更好地利用酵母所依赖的食物，即糖。
卡罗尔说，“我们发现的一个崭新东西就是基因复制。当你有一个基因的两个复制品时，有利改变可能就会出现，使一个或两个复制基因能够出现新的功能，并保留老的功能。这一现象将每时每刻在每一个活的生物体中发生。我们当中的许多人正携带有复制基因，只是我们并没有意识到这些基因复制的开始和结束。”
卡罗尔称，简尔言之，两个基因可能比一个更好，这正如多余劳动力会促使劳动力分工一样。基因能够做很多事情，复制添加一个新的遗传源，这一新遗传源可以分担工作量或新添功能。比如，就人而言，我们能够看到颜色，需要我们具有不同的分子接受器，以辨别红色和绿色的差别，但是这两种分子接受器可能都来源于相同的视觉基因。卡罗尔称观测进化进程的难点在于自然遗传改变经常是非常缓慢的，化学碱基对的微小改变就需要基因数千至数百万年的日积月累。
测量此类微小改变需要一种像简单的啤酒酵母一样的生物体模型，在相当短的一段时间内就可以产出大量后代。卡罗尔称，酵母是非常完美的，因为他们的生殖能力使遗传改变研究能够更加深入，获得更加精确的研究成果，研究人员可能生产和快速计算大量生物体的基因改变情况。如果对果蝇这种生物学最佳模型进行相同研究，则需要能够装满一个足球场的果蝇，且工作量也要多年时间。
卡罗尔称，“在非常微小的进化中，基因分裂过程往往会变得更好。当经过一段时间后，这些非常微小的变化将使一群生物体获得成功，他们将把其它生物体淘汰出局。”
新研究从不同地区的酵母中提取出基因组，以评估他们对孪生基因表现的影响力，同时对另外一个仅保留有单一基因副本的酵母基因进行处理。美国威斯康辛大学麦迪逊分校科学家解释到，“我们对进化过程进行还原”。此项研究详细演示了古代基因是如何通过复制和劳动分工提升效率的。
卡罗尔说，“他们在工作中采取最佳方式。他们共商工作，共同工作使他们的表现比古代基因的更好。自然选择使一个基因具有两种功能，创造一个两种特定基因的组装线。”
原文：
A gene divided reveals the details of natural selection
In a molecular tour de force, researchers at the University of Wisconsin-Madison have provided an exquisitely detailed picture of natural selection as it occurs at the genetic level.
Writing today in the journal Nature, Howard Hughes Medical Institute investigator Sean B. Carroll and former UW-Madison graduate student Chris Todd Hittinger document how, over many generations, a single yeast gene divides in two and parses its responsibilities to be a more efficient denizen of its environment. The work illustrates, at the most basic level, the driving force of evolution.
"This is how new capabilities arise and new functions evolve," says Carroll, one of the world's leading evolutionary biologists. "This is what goes on in butterflies and elephants and humans. It is evolution in action."
The work is important because it provides the most fundamental view of how organisms change to better adapt to their environments. It documents the workings of natural selection, the critical idea first posited by Charles Darwin where organisms accumulate random variations, and changes that enhance survival are "selected" by being genetically transmitted to future generations.
The new study replayed a set of genetic changes that occurred in a yeast 100 million or so years ago when a critical gene was duplicated and then divided its nutrient processing responsibilities to better utilize the sugars it depends on for food.
"One source of newness is gene duplication," says Carroll. "When you have two copies of a gene, useful mutations can arise that allow one or both genes to explore new functions while preserving the old function. This phenomenon is going on all the time in every living thing. Many of us are walking around with duplicate genes we're not aware of. They come and go."
In short, says Carroll, two genes can be better than one because redundancy promotes a division of labor. Genes may do more than one thing, and duplication adds a new genetic resource that can share the workload or add new functions. For example, in humans the ability to see color requires different molecular receptors to discriminate between red and green, but both arose from the same vision gene.
The difficulty, he says, in seeing the steps of evolution is that in nature genetic change typically occurs at a snail's pace, with very small increments of change among the chemical base pairs that make up genes accumulating over thousands to millions of years.
To measure such small change requires a model organism like simple brewer's yeast that produces a lot of offspring in a relatively short period of time. Yeast, Carroll argues, are perfect because their reproductive qualities enable study of genetic change at the deepest level and greatest resolution because researchers can produce and quickly count a large number of organisms. The same work in fruit flies, one of biology's most powerful models, would require "a football stadium full of flies" and years of additional work, Carroll explains.
"The process of becoming better occurs in very small steps. When compounded over time, these very small changes make one group of organisms successful and they out-compete others," according to Carroll.
The new study involved swapping out different regions of the yeast genome to assess their effects on the performance of the twin genes, as well as engineering in the gene from another species of yeast that had retained only a single copy.
"We retraced the steps of evolution," the Wisconsin biologist explains.
The work shows in great detail how the ancestral gene gained efficiency through duplication and division of labor.
"They became optimally connected in that job. They're working in cahoots, but together they are better at the job the ancestral gene held," Carroll says. "Natural selection has taken one gene with two functions and sculpted an assembly line with two specialized genes."
