26 Feb 2009

Unnatural Selection

Learning from Mother Nature, a creature or institution will thrive and survive environmental changes if it can do the followings:

1. Allow mutations or new ideas to occur.
2. Let mutants compete with wild types.
3. Accept the result of natural selection.

Unfortunately, most institutions overlook the latter two points. When a new policy comes up, it replaces the old one totally although there is no evidence that it is superior. Many people hate conservatives who reject all new policies – ward arrangement, teaching timetable, office renovation, you name it. Nevertheless, conservatives are very important. They preserve the good old days as far as possible and put the burden of proof on those who propose changes. Since bad ideas outnumber good ones in general, brutal replacement of old policies with new ones tends to result in disaster in the long run.

In the January issue of Proceedings of the National Academy of Sciences, Chris Darimont and colleagues described vividly the result of unnatural selection. Under normal selection pressure, the mortality rate of a species is usually limited, and immature and younger organisms die first. In contrast, in ecosystems affected by human predators such as hunters and fishermen, the mortality rate is higher, and the largest and mature organisms are the main targets. As a result, evolution under this unnatural selection is much faster. Interestingly, the size of organisms decreases by an average of 20%, and the life cycle shortens by nearly 25%. The poor preys just have to reproduce earlier to avoid extinction.

Sadly, the result of this report reminds us of ourselves. Good clinical research takes years or even decades of careful observation. Under the current method of research assessment, academics who only do big and long-term projects will die first because of insufficient research output. Next time when you read a me-too study or non-definitive study reporting dubious surrogate end points, don’t laugh. Those people are just shortening their reproductive cycle because of unnatural selection!

19 Feb 2009


Between 1405 and 1433, Zheng He of the Ming Dynasty led a fleet of more than 20,000 men to explore the ocean west to China. The countries he visited included Indonesia, Yemen, Iran and Somalia in East Africa. The largest ship could carry 200 sailors and 1000 passengers. This took place before Christopher Columbus was born (1451-1506). The fleet was also much larger than the one led by Columbus. Although Zheng He’s mission was never meant to be colonization, we can still expect that the international influence of China would grow with time if similar voyages continued.

However, the policy of the Ming Dynasty changed suddenly afterwards. The country not only abandoned further sea adventures, but also cut off most international interactions. A number of hypotheses explained the change. Some said that the emperor used up all the finest lumber to build the Forbidden City. Others said that the main reason was the political rivalry between eunuchs and the royal family (Zheng He was a eunuch).

While we consider the abandonment as one of the foolish decisions in Chinese history, one may wonder why the same thing did not happen in Europe. New ideas are like mutations. If the idea of going out to search for unknown land (Columbus’ calculations and predictions before his voyage were almost completely incorrect) is a mutant, the wild-type would be minding your own business. In fact, when Columbus approached John II, King of Portugal, for funding, he hesitated. That is why Columbus had to turn to the Spanish Queen. The interests in sea adventures also fluctuated among different European countries.

What set Europe apart from China was the keen competition among countries. The great success of the Spanish campaign brought in gold and resources, building Spain into an empire in the 16th century. Other countries sensed the selection pressure and had to follow suit. This resulted in a long history of colonization. In contrast, China did not experience major economic and political competition from neighboring countries, and could live contently on her own until the European navies knocked her door.

12 Feb 2009


New ideas are like mutations. In Built to Last, James Collins and Jerry Porras illustrated how companies that encouraged new ideas outperformed those that explicitly avoided changes. The mutations allow the companies to change over time and face up new challenges. In Chinese history, many dynasties upheld the belief that ancestral decrees should never be disobeyed, and headed towards ruin when the environment changed.

Therefore, I cannot fully support those who reject all new policies such as the education reform and ward admission system. Of course, this is not to say that the results of these new policies are gratifying. In fact they aren’t. Supporters of conservative approach may point out that though the dynasties collapsed, they had each lasted for several centuries. In contrast, Wang Mang (王莽) established the Xin Dynasty (新朝) and was well known as a creative politician and scholar. He was described as a ruler who announced a decree in the morning and changed it in the evening (朝令夕改). The empire only lasted from 9 to 23 AD. In Fooled by Randomness, Nassim Nicholas Taleb explained the cost of progress (or change):

‘People tend to infer that because some inventions have revolutionized our lives that inventions are good to endorse and we should favour the new over the old. I take the opposite view. The opportunity cost of missing a “new new thing” like the airplane and the automobile is minuscule compared to the toxicity of all the garbage one has to go through to get to these jewels (assuming these have brought some improvement to our lives, which I frequently doubt).’

This leads us to fundamental questions – Does the outcome depend on the quality of the mutations? Is there an optimal mutation rate?

If we turn to Nature for an answer, the first question is obsolete. Mutations are random processes. History tells us that bad ideas outnumber good ones. Great minds do not guarantee better ideas. We only know that bad ideas by great people usually result in worse disasters (think about Karl Marx). So we have to accept the co-existence of good and bad ideas.

Similarly, microorganisms with the highest mutation rate are the most robust. HIV, hepatitis B and hepatitis C viruses can all produce mutations at any point in the genome within one day. They are also very successful in developing drug resistant mutants.

Then what went wrong? Why do the government and some institutions do badly even though the officials bombard us with innovative ideas all the time?

5 Feb 2009


My colleagues were surprised when I celebrated the 200th anniversary of the birth of Charles Darwin in a public talk on hepatitis B recently.

Explaining the evolutionary theory to non-specialists is a bit tricky, but people working on antiviral therapy use the theory all the time. The essence of the evolutionary theory is ‘survival of the fittest’. During reproduction, gene mutations sometimes result in phenotypic changes in an organism. Most of the mutations are hazardous, but a minority confers competitiveness. Organisms bearing more competitive mutations have a better chance of giving birth to offsprings, and eventually the new gene will dominate the population. People working with viruses are seeing a fast-forward version of evolution. Viruses replicate quickly and produce a lot of mutations because of their lack of proof-reading during gene transcription. When an antiviral drug is given to a patient, viral mutants that do not fear the drug will replicate more effectively and soon become the dominant strain.

Actually, before Darwin published his book On the Origin of Species, a few other scientists had already noticed the gradual phenotypic changes of animals and plants. However, most considered the changes a well-planned evolution by God. It was Darwin who spotted that the changes were random, and the evolution was driven by natural selection – ‘survival of the fittest’. For those with basic knowledge on genetics, the theory may seem pretty easy to understand. However, one must remember that genetics did not exist at Darwin’s time (although Gregor Mendel was actually humbly doing his cross-breeding experiments at roughly the same time). Imagine how difficult it would be for me to explain the last paragraph again without describing anything about genes and mutations.

Like all great theories, the implications extend well beyond the original territories. I will elaborate this further next week.