Complexity Digest 1999:beta6
(Archive: http://www.comdig.org, Mirror: http://www.comdig.de/)

 

1. How Neurons TeamUp To Recognize Patterns

One of the big open questions for the next millennium is how the brain uses neurons to represent decisions, perceptions, emotions and all the other things we can do with our brains. Some claim that it is impossible at all that neuronal activity alone can explain consciousness.

After the discovery of neurons with very specific response to very specific sensory input (e.g. different images that somehow represent a person looking at something on the floor) the existence of "grandmother neurons" was postulated: there exist a single neuron (out of the tens to hundreds of billions of neurons) in the brain that will become active only if you see (or hear or think of) your grandmother. That view is similar to that of the brain as a digital computer where all bits of information are stored at well-defined addresses on a chip. Experimental evidence does not support this position. For instance Walter Freeman could show that in the area of the rabbit brain that responds to different odors that there are not different neurons for each smell. Instead different collections (assemblies) of neurons become active every time the animal smells a familiar odor. It seems that these assemblies themselves change over the course of time. Note, however, that all of these effects are very small and buried in the chaos of spontaneous neuronal activity.

A group of scientist at the Weizman institute in Israel has developed a technique to visualize brain activity directly with the help of dyes that change their color under the influence of the tiny electrical fields that are produced by the brain. In this way they could observe the formation of neuronal cell assemblies in cats in response to moving optical stimuli. In their current paper they undergo a very detailed and careful data analysis to link the activity of individual neurons to that of the cell assembly. They could very elegantly establish a "Preferred Cortical State (PCS)" for each neuron studied. It is defined as an activity state of the brain (or neuronal assembly) for which the given neuron shows maximum activity; it is in resonance with its assembly.

The consequences of this finding is fascinating: For instance Tsodyks et al. could show that brain states similar to the neuron's PCS are generated intermittently and the individual neuron will respond with an enhanced activity with or without a stimulus. In a way it seems that the brain is spontaneously browsing through a number of (expected?) states and if the external stimulus actually happens it can rapidly switch to the new state.

This phenomenon of "readiness-by-anticipation" is known for the auditory system. It makes evolutionary sense because it shortens the response time to visual stimuli and that can be crucial for the question of eating or being eaten.

Linking Spontaneous Activity of Single Cortical Neurons and the Underlying Functional Architecture, M. Tsodyks, T. Kenet, A. Grinvald, A. Arieli, Science, Volume 286, Number 5446 Issue of 3 Dec 1999, pp. 1943 - 1946

2. Sea Ice Melt-Down And Climate Change

The earth climate system is certainly one of the complex systems on our planet with the most severe impact on our lives. The complex interactions of the atmosphere with factors like the oceans, ice-shields, and of course human activity leads to and modifies self-organized, large-scale coherent patterns like hurricanes and the El Nino oscillation. Several articles in different journals discuss some of the recent findings into the complexity of climate dynamics. One of the most dramatic and erratic change patterns of terrestrial climate are the occurrence of ice ages in the northern hemisphere. There are a number of factors that can cause transitions between ice ages and inter-glacial periods. One of them could be the Arctic Oscillation (AO) an erratic (chaotic?) oscillation of atmospheric pressure systems over the North Pole. Another self-organized structure with significant importance especially for the climate in Northern Europe is the "North Atlantic" conveyor belt. It is primarily powered by cold seawater that is formed in the North Atlantic. It because it is more dense it sinks to the bottom of the ocean. The displaced water forms a deep ocean current going South. There it displaces warm tropical surface water that flows back up North (e.g. as the "Gulf Stream") and closes the circle. If for some reason the conveyor belt slows down it will cause a significant drop in temperature in the North and a raise in tropical sea-surface temperatures in the South. The resulting larger temperature difference in the atmosphere will in turn create wind and cloud patterns that can either weaken (negative feedback) or strengthen (positive feedback) of the weakened conveyor belt.

The team of C. Ruehlemann et al. did a careful investigation into the history of past sea surface temperatures in the Northern Atlantic. They wanted to find out if climate changes happened uniformly over the Atlantic or if there was indeed a cooling in the North while there was a warming in the South. The first option they identified with greenhouse gases the second one with a breakdown of the conveyor belt. They could provide convincing evidence that indeed the conveyor belt action was a good indicator for climate change. Is this an argument against greenhouse gases as contributing factors for climate change? As the primary cause that makes the conveyor belt stop the authors mention fresh surface water in the North Atlantic. Since cold fresh water is not as heavy as cold salt water the surface water will not sink to the bottom of the ocean and the primary driving force has vanished.

C. Ruehlemann et al. did not discuss where the fresh surface water was coming from. R.A. Kerr mentions that runoff from melting sea ice could be one large freshwater source. Sea ice has been melting at the alarming rate of 15% per decade. Arctic ice has lost 40% of its volume over the past 30 years. Instead of an average thickness of 3.1 meter arctic ice is currently down to about 1.6 meter. Changes in the temperature could have caused changes in storm tracks, therefore changes in cloud coverage and therefore changes in precipitation patterns. Which in turn will affect the sea ice runoff.

K.Y.Vinnikov et al. have done large-scale computer simulations of the chaotic fluctuations of climate change over a period of 5000 years. They conclude from their results that the chance that the current sea ice meltdown is just a large natural swing that will correct itself after a while is 1:1000.

Therefore it is predictable when at the current rate all arctic ice cover will be gone. The fourfold change in albedo from white ice to black water will induce a correspondingly increased heating of the surface water in Polar Regions. This will in turn reduce the mixing of surface and deep water, which means that the oceans will not act as CO2 and heat reservoir.

All these complex feedback loops are classical ingredients of complex systems. The challenge will be to apply methods from complexity theory to attack the problem. Will control of chaos applied to geo-engineering help to get the conveyor belt jump-started again?

1. Arctic Thawing May Jolt Sea's Climate Belt, William K. Stevens, NY Times, December 7, 1999

2. Will the Arctic Ocean Lose All Its Ice? Richard A. Kerr, Science,  Volume 286, Number 5446 Issue of 3 Dec 1999, p 1828,

3. Global Warming and Northern Hemisphere Sea Ice Extent, Konstantin Y. Vinnikov, Alan Robock, Ronald J. Stouffer, John E. Walsh, Claire L. Parkinson, Donald J. Cavalieri, John F. B. Mitchell, Donald Garrett, Victor F. Zakharov, Science, Volume 286, Number 5446 Issue of 3 Dec 1999, pp. 1934 - 1937,

4. Warming of the tropical Atlantic Ocean and slowdown of thermohaline circulation during the last deglaciation Carsten Rühlemann, Stefan Mulitza, Peter J. Müller, Gerold Wefer & Rainer Zahn, Nature 402, 511 - 514 (1999),

3. Predicting the Next Move of Flu Viruses

Biology does not have the reputation of being a terribly predictive science. Arrogant physicists even joke that biological research consists of replacing a mystery by a miracle. But biological systems are intrinsically complex since they typically evolve and adapt. And me know already from simple, non-linear, iterative function systems and cellular automata that it is basically impossible to predict how the behavior of a system will change in response to a modification of its dynamical rules.

One of the most visible examples of this problem is the common flu: Apparently it is easier to send a man to the moon than finding a cure for the common cold. While the moon doesn't try to hide or change its orbit the influenza virus developed a whole bag of tricks to fool our immune system.

For instance the influenza A virus has developed a stealth technology to hide from our immune system so that it has a greater chance to spread every flu season before it gets detected. Also immunization based on last year's flu is largely ineffective because the virus has already mutated to a new form that is not recognized by the immune system. Bush and Fitch studied the mechanism by which influenza A is doing this trick. The immune system recognizes the virus with the help of a protein (hemagglutinin), on the surface of the virus. By comparing different virus generations over a number of years the researchers could recognize a pattern in how the successful viruses mutated. They tested their hypothesis over eleven flu seasons and came out with a highly significant match. That means they have now a predictive tool to anticipate the virus' next move. That does not imply that they know what strain of viruses will hit in the next flu season (certain strains persist for several years) but at least they have some idea what sort of vaccines to develop to be ready for the next influenza mutation. It would be interesting to see if in this biological arms race the virus will be able to adapt to this move of human science and change its mutation patterns.

1.Predicting the Evolution of Human Influenza A, Robin M. Bush, 1* Catherine A. Bender, 2 Kanta Subbarao, 2 Nancy J. Cox, 2 Walter M. Fitch 1, Science, Volume 286, Number 5446 Issue of 3 Dec 1999, pp. 1921 - 1925

2. Predictive Evolution, David M. Hillis, Science, Volume 286, Number 5446 Issue of 3 Dec 1999, pp. 1866 - 1867,

4. "Chaos" Theory Empowers VA And University Of Florida Researchers To Predict Epileptic Seizures (Excerpts from University Of Florida Health Science Center press release)

Inspired by an intriguing mathematical concept known as chaos theory, researchers at the University of Florida Brain Institute and the Malcom Randall Veterans Affairs Medical Center in Gainesville have developed a technique for predicting some types of epileptic seizures minutes to hours before they begin.

Their work, under development for more than a decade and now the basis of a U.S. patent application, opens the door to the creation of implantable devices that can detect signs a seizure is approaching and deliver medication, or electrical or magnetic stimulation to try to prevent it.

"We had determined some years ago that there was a theoretical potential for predicting seizures," said Dr. J. Chris Sackellares, a VA neurologist and a UF professor of neurology, neuroscience and biomedical engineering who has been researching temporal lobe epilepsy with Leonidas D. Iasemidis, a VA research engineer and UF research assistant professor of electrical and computer engineering, and neuroscience. "But it has only been in the past year that we have really been able to demonstrate that we can do so reliably. And it has only been recently that we realized that the state of transition to a seizure could last for many hours."

(…)

The researchers began to suspect in 1988 that the emerging field of chaos science would be able to shed light on epilepsy. In a nutshell, chaos theory offers a mathematical approach for seeing a kind of order in events that previously had appeared to be random. Early in the 1990s, the approach enabled Sackellares and Iasemidis to be the first to identify the existence of a pre-seizure transition period.

In their early research the scientists were looking for such transitions to occur seconds to minutes before a seizure began. But in the past year, by analyzing electrical activity in the brain recorded for a 10-day period, they have identified a warning stage developing anywhere from minutes to many hours ahead of time.

Their technique involves using sophisticated mathematical formulas to sort through the brain's complex electrical signals, which can be recorded by electroencephalograms, or EEGs. The scientists theorize that a seizure's function is to correct a neural system gone awry. Though it may sound counterintuitive, a buildup of organized, harmonious signals is what apparently needs to be fixed to return the brain to its naturally chaotic state.

To predict seizures, Iasemidis, who directs the Gainesville VA's Brain Dynamics Laboratory, and Sackellares look for signs of communication between the site where a patient's seizure begins and elsewhere in the brain. When an increasing number of electrode pairs begin oscillating together during an EEG, it signals a seizure is on its way.

(…)

Iasemidis noted that their goal is to be able to identify a window of opportunity for preventing seizures. "Predicting exactly when a seizure will occur is not the main question," he said. "We're interested to see if we can knock the system out of its route to the seizure. We'd like to see if we can intervene with either electricity or medication to try to get the system to reset itself right at the beginning of the buildup of the pre-seizure transition.

(…)

But Sackellares and Iasemidis say it is realistic to think that implantable devices can be developed to detect the preseizure state and automatically act to thwart it. They noted that such devices have been developed for other conditions, including diabetes.

A scientist who collaborated with Sackellares and Iasemidis in the early 1990s said he had been skeptical back then that the computational difficulties of the line of research could be overcome.

"It seems that they have been able to press ahead and realize their long time-dream of making the methods associated with chaos theory practical," said William J. Williams, a professor of electrical and biomedical engineering and computer science at the University of Michigan. "No one else has pursued this direction of research so persistently in order to achieve such an advanced understanding of epilepsy. The practical applications of their work will likely have a lasting beneficial effect on many people who suffer from epilepsy.

University Of Florida Health Science Center, Victoria White , Office Of Public Information

5. Another Miraculous Property of Water, Sander Woutersen et al, Nature

It is no accident that researchers look for signs of water on other planets to determine the possibility for life. Its amazing properties make water indeed a very special liquid. Woutersen and Bakker added another miracle property to the list that confirms that water is indeed the perfect environment for life-like processes. They could show that water can act as extremely fast conductor of energetic states between and within molecules especially bio-molecules. Many organic molecules have OH groups (i.e. pairs of an oxygen atom connected to a hydrogen) attached to them. Their structure and state determine many of the chemical properties of the molecule.

The OH groups also acts as a springs that can store energy in the form of stretch-vibrations. Exactly which of the many OH groups in a large bio-molecule are in such a vibration mode and how much energy is stored determines which reactions will take place and which ones will be suppressed. Therefore the fast and effective transfer of energy between different OH vibrations can be crucial for many of the central processes in living systems. Once excited an OH vibration looses its energy already after about 740 femto-seconds. That is more than a thousand times shorter than the time it takes the fastest Pentium chip to complete one elementary operation. During that time the OH spring oscillates less than a hundred times.

With the help of different mixtures of light and heavy water the researchers could demonstrate the because of the special properties of water it can transfer the energy stored in these vibrations fast enough to prevent it from being dissipated. Although the experimental evidence is quite convincing that the energy transfer happens it is still not clear how exactly this works. It seems clear, however, that non-linear resonance phenomena are essential in this process.

Resonant intermolecular transfer of vibrational energy in liquid water, Sander Woutersen, Huib J. Bakker, Nature 402, 507 - 509 (1999)

6. Physicists and Astronomers Prepare for a Data Flood, Mark Sincell,Science

The spread of the Internet spawned an increasing number of publications that speculate about a new evolutionary level of organization of humans, networked together via the Internet. One question naturally comes up about the need for such a superstructure. What problems could be solved that cannot be solved by traditional team working?

The paper by Mark Sincell might provide one answer: A new generation of data-recording devices in science and especially in particle physics and astronomy will provide a stream (better: torrent) of information that goes orders of magnitudes beyond ordinary human experiences. The planned Large Hadron Collider at CERN in Geneva will write data to a disk-based database at a rate of 100MB per second up to a total of 100 petabytes = 1017 bytes that is the equivalent of about 10 million feature length movies (in DVD resolution). Or, according to Arthur C. Clark's science fiction novel "3001", the totality of all experiences of a human being during the whole life span i.e. an old person's brain content if that person had perfect memory.

Although particle physics is notorious for collecting tons of data in order to detect a handful of "interesting events" -like the trace of a Higgs particle- this example demonstrates the domains of information processing that will be happening in the next millennium.

Astronomers are expected to collect data about 200 million galaxies in the amount of a mere 40 Terabytes. They are also known to actually do something with their data like collecting statistics and detecting patterns in their positions and motions. Storing the data at a central location and downloading them to individual users (today's standard client-server model) is not feasible for that amount of data. A new, hierarchical structure in the data storage organization will move data closer to the user in several tiers depending on the actual use patterns. The concept of linking data storage with usage is not new: memories in the brain are stored in a way that critically depends on usage frequencies.

Because of the globally accessible enormous quantities of information some researchers predict that new discoveries will emerge with the help of automated search software that would have been beyond the capabilities of humans.

Physicists and Astronomers Prepare for a Data Flood, Mark Sincell, Science, Volume 286, Number 5446 Issue of 3 Dec 1999, pp. 1840 - 1841

7. The DNA sequence of human chromosome 22, I. Dunham et al, Nature

The human genome project is to biology what high energy physics is to physics: Big Science where the length of the list of authors of becomes a major fraction of the publication of the research results.

To decode the software that assembles us humans is an ongoing global collaboration that predictably will be finished in the very early part of the third millennium. This predictability of the research results is also a feature common with elementary particle physics where one can predict the date at which enough events will be collected to be published in the first rate journals.

I. Dunham and 215 collaborators succeeded in decoding chromosome No. 22 (out of 46) consisting of more than 33 million bases, 545 genes, and 134 pseudogenes. It is the first complete mapping of a human chromosome which covers almost 2% of the genomic DNA.

This research group used the "clone-by-clone" approach of two commonly used methods to sequence the human DNA. It covers piece by piece parts of the DNA and determines its complete sequence. The alternative is a "shotgun approach" that sequences much smaller segments selected from the whole genome.

There are some medical reasons for starting with chromosome 22: It is the location for a number of human congenital anomaly disorders including cat eye syndrome and the schizophrenia susceptibility locus.

The authors call their results an " operationally complete genomic sequence of a chromosome" but admit that they were unable to obtain sequence over 11 small gaps using the available cloning systems but express hope that these gaps will be closed with improved methods.

All the results are continuously made available to the public via the Internet: The complete sequence and analysis is available at (http://www.sanger.ac.uk/HGP/Chr22 and http://www.genome.ou.edu/Chr22.html).

The DNA sequence of human chromosome 22, I. Dunham et al, Nature 402, 489 - 495 (1999)

 

8. The Unexpected Science to Come,  Sir John Maddox, Scientific American

The last Scientific American issue of 1999 is dedicated to a discussion what kind of discoveries can be expected in the first 50 years of the new millennium. It is quite adequate that the former editor of one of the most prestigious (and oldest) science journals of the second millennium, Sir John Maddox of Nature, gives both a review and a preview of the most fundamental discoveries in science.

One of the easier forecasts is the completion of the human genome project with its tremendous implications not only for medicine but also for the reconstruction of the genetic reconstruction of the human race. Will it be possible to explain the transition from great apes (having 48 chromosomes) to humans (only 46 chromosomes) and why only a single species of humans survived?

Maddox sees a second evolution related fundamental question in the interpretation of the non-coding "junk" parts of the DNA. It could shed some light on living things in the "RNA world" that supposedly preceded the "DNA world" we are living in today. He does not expect, however, that anybody will be able to build a complete RNA organism within the next 50 years.

Another open fundamental question is that of the origin of life. Maddox expect that a better understanding of how solar radiation can interact with molecular cloud in outer space to form increasingly complex molecules including fullerenes (commonly known as "buckyballs").

He also expects that biology as a science will become more predictive and quantitative instead beyond discovering more "miracle" enzymes with peculiar functions. For instance the cell-division cycle has been the subject for discovering new enzymes at a rate of one new enzyme per week but without answering the questions about how the enzymes which trigger the cell cycle are triggered.

There is hope that some of the fundamental questions of brain science will be answered in the next fifty years, for instance which neuronal processes correspond to the act of making a decision.

The Unexpected Science to Come,  Sir John Maddox, Scientific American, Special Issue: What Science will know in 2050, 12/99

9. Neanderthals Worked Themselves to Death

If we take brain size as a measure for a species' position on the evolutionary ladder then two completely different candidates for the top clearly stand out: humans and cetaceans. One of the most puzzling differences is that only a single species of humanoids survived a mere few million years of evolution after they separated from their closest ancestors, the great apes. Cetaceans, on the other hand, have branched into a large number of different species, from small dolphins to the largest mammal that ever inhabited this planet, the Blue Whale. These numerous species by no means peaceful creatures: dolphins gang up against porpoises and kill them; killer whales hunt and kill other whales. But in spite of this they have been able to coexist for tens of millions of years in basically their current form.

The last surviving sister species of homo sapiens, the Neanderthals went extinct just a little over 20,000 years ago after having adapted to challenging climate changes over a period of 250,000 years and after coexisting with modern humans for thousands of years.

Duke University anthropologist Steven Churchill thinks that ultimately, the physical costs of their hunting methods may have been too high for Neanderthals to survive. Their anatomy was supposed to be less efficient that that of Cro-Magnons for hunting large animals. Furthermore they might have lacked the language and strategic planning skills to prevent extensive periods of famine.

Although these hypotheses have a level of plausibility it seems unlikely that they fully explain the mystery of the disappearing Neanderthals.

Neanderthals Worked Themselves to Death, Blake Edgar, Discovery News Brief , 5 Dec. 1999

See also Youngest Neanderthals Yet Found, The Fate of Neandertals

 

10.  IBM’s new supercomputer, CNN/Reuters

IBM's claim to fame in the area of super-computing in the past was not so much based on scientific discoveries or simulations but more in the area of finding the biggest prime numbers and beating human chess grand-masters.

This might change with the latest IBM supercomputer -named Blue Gene- that was announced to become fully operational in about five years. As the name suggests its design purpose will be in one of the areas of complex systems research: protein folding (see ComDig 99:beta4.1).

The strategy that IBM follows to achieve a one quadrillion operations per second (two million times the speed of the fastest microcomputer today) is not to develop faster processors but to use current off-the shelf processors and use it in a massively parallel way.

It is known that this route requires special demands on software as well as on the problems for which this speed-up can be achieved. For instance the time to multiply two numbers will not be much faster on Blue Gene than on four notebook computer. But for problems like protein folding where the movement of a large number of atoms have to be tracked simultaneously it will be possible to distribute the computational load from the movement of different atoms to different processors.

IBM’s new supercomputer, CNN/Reuters, December 06, 1999: 7:45 a.m. ET

11.  Ancient origins of nitric oxide signaling in biological systems, Jörg Durner et al, PNAS

Complex adaptive systems need efficient communication among their sub-units or agents such as chemical trails in ant colonies and electrical nerve pulses in neuronal systems and brains. Just like parts of the brain show their ancient evolutionary history it seems that also the signaling methods in organisms have inherited some pathway from ancient evolutionary times. Whereas it has been known for a long time that organisms use hormones (relatively complex proteins) to regulate body functions and emotional responses it is less known that the simple molecule nitric oxide (NO, consisting only of the two atoms nitrogen and oxygen, both abundant in our atmosphere) plays an astounding signaling role in different systems of our body. For instance it helps to keep our blood pressure constant, to stop bleeding, and even to transmit nerve signals in a chemical and not electrical manner. It also plays an important role in our immune system by contributing to the activation of cellular defenses against pathogens. Surprisingly it also is involved in a similar function for plants. Furthermore NO helps the blood hemoglobin to deliver oxygen to tissue that is in need of it. It does that by causing blood vessels to dilate and thereby increase the supply with oxygen-rich hemoglobin. Durner et al. were able to analyze in great detail the chemical processes that make these functions possible. The importance of NO for controlling biological functions is also supported by the fact that a special enzyme exists that will produce new NO in animals. In spite of the functional similarities with plants, however, it was not possible so far to also identify a similar enzyme produced by plants.
Ancient origins of nitric oxide signaling in biological systems,  Jörg Durner, Andrew J. Gow, Jonathan S. Stamler, and Jane Glazebrook , PNAS, Vol. 96, Issue 25, 14206-14207, December 7, 1999

12. Immortality and Cancer, B.-S. Herbert et al, PNAS

The age of a cell can be determined from the length of its telomeres (the end section of a chromosome, composed of several hundred base pairs): Every cell division shortens the telomeres until cell undergo apoptosis and die. The enzyme telomerase can lengthen the telomeres again and extend cell life theoretically making them immortal.

Using telomerase as a life prolonging agent has its risks: It is known that telomerase is especially active in cancer cells basically allowing them to continue dividing without limit. Herbert et al. studied how agents that inhibit can indeed cause cancer cells to age and eventually die. If telomerase is activated for aging cancer cells the process is reverted, the telomeres grow back to their original length and the cell is immortal again.

The telomere age of cells has an interesting consequence for cloning: It turns out that clone organisms are born with the cell-age of the donor organism. That puts a damper on the hope that cloning could be used to achieve immortality: the cells of the cloned baby are just as old as those of the parent.

Inhibition of human telomerase in immortal human cells leads to progressive telomere shortening and cell death, B.-S. Herbert, A. E. Pitts, S. I. Baker, S. E. Hamilton, W E. Wright, J. W. Shay, and D. R. Corey, PNAS, Vol. 96, Issue 25, 14276-14281, December 7, 1999 

ComDig 1999:beta6

10-Dec-1999