Wednesday, February 27, 2013

The Complexity Lab Manual View Of What is Life and How Can It Develop Spontaneously From Chemistry

Diversity and Complexity of Life

1.1) you can go outside in almost any noncity street and collect 100 different kinds of plants.  Along the way you will get some experience with diversity (almost continuous variation on a theme), disparity (discontinuity between forms), and the incredible morphological details that allow us to notice these differences

1.2) not only morphological details, but behavioral.  You can watch a beehive and eventually learn that a 2cm long honeybee can execute over 270 different skills, one of which is to reproduce new elaborate honeybees out of the food they search for and collect.  How do they do this?  Can we build robots to do this?

Computer Science

To get a handle on the levels of complexity involved in life we can get hands on experience building complexity from the ground up.

2.1) from transistors to logic gates, from logic gates to flip flops and multiplexers, from these elements to memories and arithmetic processing units, from these structures to computers with instruction sets of on the order of dozens of instructions.

2.2) once we have computers we can create wild complexity by combining these instructions again hierarchically into subroutines of subroutines ... building up programs a billion instructions long.

2.3) we can even use these programmed computers to hook up to sensors and actuators of robots.

So, can we build 2cm autonomous self feeding reproducing Honeybee robots?  No! not yet. and the reproducing part we still haven't a clue about!  But we've learned alot along the way.

Animal and Plant Development

So, how do animals build their children out of food?

We can watch an animal develop from its egg.  we learn that an animal is actually a growing colony of self reproducing single celled organisms.  a Honey bee starts off it's life as a single celled egg/amoeba that splits in half to 2 then 4 then 8 500million swarming self organizing amoebas that remain connected and make a honeybee happen. (actually teh first 10 divisions to 1024 cells are slightly different than this)

3.1) Pretty much everything that most organisms can do, this basic unit of life: the cell, can do.  dozens of behaviors, food seeking, reproducing itself out of food.  We can watch single celled Paramecium, Stentor, Euglena go about their ways in a drop of pond water under the microscope.  With months of observation and experiments we can catch them at many of these behaviors.

So What Is This Basic Unit Of Life, The Cell?

4.1) a 150 years of at first confusing but slowly clarifying chemistry experiments has shown us that we can grind up cells and separate them out to 100s of distinct kinds of small components and 1000s of distinct kinds of large components built of these small components.  eventually we learn that the simplest of cells, an E. coli bacteria, are swirling underwater cities of  3000 protein building machines building 2million protein worker robots swimming around and being taken apart and rebuilt out of a soup of a billion parts in a sea of 10s of billions of water molecules.

4.2) I compare this to a city because i can approximate how many bricks there are in New York City:

100bricks per window *100windows per wall
or 100^3 bricks per building * 5x10 buildings per city block,  *10*200 city blocks in Manhattan *5 boroughs in all of new york city
=10^10x5=50billion, yeah more atoms in an E. coli bacteria than there are bricks in new york city.

4.3) remember in our computer chapter we learned what kind of advanced behaviors we could program out of a billion parts!  but there is a profound difference between molecular biology of the cell and programmable computers:  a computer program might consist of a million subroutines and state variables, but the computer processing unit processes these ONE AT A TIME or at most with some parallelism a dozen at a time.  In a living cell, those 2million proteins equivelent to simple subroutines are ALL OPERATING IN PARALLEL, It's like having a million different computer chips interacting at once.  and they dont have to execute the billion instructions of the program, because the billion parts are also bouncing around and interacting in parallel on their own.

so a single cell is a whole qualitatively different level of complexity than a single computer running a program.

4.4) Notice I haven't discussed DNA.  at this level of discussion it doesn't add much important difference.  The 10,000 genes in a cell are simply another 10,000 subroutines running in parallel with and interacting with the 1000s of different kinds of protein robot/subroutines. [if you are into object oriented programming, you can think of them as the classes and the proteins as the objects generated by them, which of course can use the classes to generate even more objects]

What On Earth Are These Molecular Parts?

5.05) they are constructed out of a finite set of only 2dozen distinct kinds of atoms.  The major players: Carbon, Hydrogen, Nitrogen, Oxygen, Phosphorus, Sulfur, Sodium, Potassium, Magnesium, Calcium, Chlorine.  The trace elements (but essential nonetheless) Iron, Chromium, Manganese, Cobalt, Nickel, Copper, Zinc, Molybdenum, Iodine.  (a few other elements are used by various organisms that aren't people)

5.1) these parts are always in 3 dimensional motion (swimming around in a sea if you will) the smaller peices vibrating and bumping into each other a trillion times a second and the larger ones a billion times a second.  this motion we get for free simply because the universe is a warm place.  this is what heat is, the motion of atoms and molecules.  this is also why the higher temperature we raise a system, the faster (usually) chemical reactions will take place.

we can see hints of this motion in the microscope, this is called brownian motion.  Einstein and Perrin showed us how to use these observations to calculate that there are those 100billion molecules in a cell.

5.2) they are sensitive to each other and their environment (pH, hydrophobic/hydrophilic phase etc..) so that each molecule has functional parts that can react to each other with 100s of specific rules.  One learns this in an organic chemistry class.

5.3) they spontaneously join with each other and split apart in different ways depending on these rules.  This is mediated by energy flow which is the topic of chapter 6.  No outside mechanic is required.

5.4) even without energy flow, at thermodynamic equilibrium due to their incessant motion, these parts can self assemble into larger complex structures because of the way they attract each other (like magnets) and simple mathematics.

5.5) one example of this is the formation of clathrin coated pits which cells use to engulf molecular packets around them and bring them inside the cell.

5.6) You can see things like this happen by watching various feathery patterns of ice form in winter.

5.7) or by looking at an exhibit of 100s of different mineral formations.

5.8) so we have these billions of parts randomly exploring each other and then reacting with specific rules.

Energy Flow

living cells are systems of molecules which have energy flowing through them.

Ultimately from the 5000 degree photosphere of the sun spewing out to the 2degrees coldness of outer space.  In between plants create high energy bonds in food molecules and these release energy to the next high energy bonds etc.. till the final fermentation processes of decay release heat to the cold night sky.  This flow of energy in a system of interacting parts can:

6.1) Energy flow causes cyclic iterations in systems of interacting parts. We can build a simple steam engine that cycles off of the flow of energy of burning fuel to CO2 and heat.

It will be important to notice that it is not JUST the heat that makes a steam engine work.  if we put the whole engine INSIDE a hot oven, it will stop.  The heat source and the Cold sink are both required.  Energy must flow from high potential to low.

6.2) Energy flow creates stable dynamic patterns in systems of many many interacting parts (fluids).  We can see this in simple heat flow creating stable patterns of convection cells in fluids and the flow of chemical energy creating stable patterns of spiral waves from a homogeneous mixture of just 5 simple chemicals in the Belusov Zhabotinsky reaction.  We can combine heat and energy flow and play with a candle flame.

6.3) the patterns are stable and can damp out certain classes of perturbation

6.4) but can also preferentially amplify other classes of random fluctuations.  for instance the spiral waves in the BZ reaction are just such amplifications of random fluctuations in the otherwise homogeneous solution.

6.5) interesting unpredictable dynamics can even be created.  We can see this by building a simple chaotic water wheel.  The BZ reaction also can run in a chaotic oscillating mode.

6.6) It is important to note that the ability of life to grow in complexity does not contradict the second law of thermodynamics.  This is true both at the level of a single organism growing from an egg/seed or at the level of whole ecosystems on earth evolving more elaborate different species and interactions between species.

[Explanation coming soon]

Emergent Complexity In Mathematical Systems

7) systems of billions of simple connected agents with simple rules of interaction, under iteration can create VERY ELABORATE patterns that keep growing.

We are just beginning to explore the world at the atomic scale of pattern with scanning tunneling atomic force microscopes.  but we can go back to our computers and play some mathematical games to see what happens when we let 1000s or millions of interacting simple components iterate their simple rules of interaction over and over again.

This is the newest area of study in this exploration!  Only 40 years old!

7.1) the first hint of this kind of behavior was John Horton Conway's game of life, which he introduced in the early 1970s.   Begin with a grid of squares each of which can either be on or off.  each square only interacts with the 8 squares around itself.  every iteration we update all the squares in parallel with these two simple rules:

1) if an on square has less than 2 or more than 3 on neighbors it turns off, else it stays on

2) if an off square has exactly 3 on neighbors, it turns on, else it stays off.

O is off, X is on

you can work out for yourself what these four patterns do:





be prepared to be amazed.

we can write a computer program to mechanize this (and can imagine one day being able to use nanotechnology to build a grid of molecules to execute these patterns!)

and with that power learn that we can find patterns that continue to grow forever, and we can even make a pattern that forever spits out a sequence of prime numbers so that it grows for.ever and gets more and more different.

But Conway life is not a very satisfying example.  mess around with it and you see that if you change ONE little square in an elaborate pattern the whole dynamic pattern can crash, it is not very robust, the way chemistry is.

7.2) an even simpler rule generates more stable complexity.  Langton's ant: again start with an empty square grid.  Langton's ant is a simple machine (again we can imagine building a simple nanomolecular machine to do this in the not too distant future)  (The more exciting prospect is that perhaps with advanced tools of observation we can discover conditions under which an already existing molecule can do something like this) with a simple rule: the machine sits on a square and faces one of the 4 cardinal directions (N,E,S,W).  The rule is: if the machine is on an off square, turn it on, and rotate 90degrees clockwise, or if it is on an on square, turn it off and rotate 90degrees counterclockwise.  then move forward one square and repeat.

Operating with this simple rule, Langton's ant generates an increasingly complex weird shape that slowly grows to about 100x100 squares over a period of about 10,300 steps and then switches to a periodic behavior whereby it travels back and forth diagonally with one move forward each time in a period of about 100 steps each back and forth and builds an infinite repeated patterned highway shooting off in one direction.  It is very stable, no matter how much you mess up its pattern, it will eventually build its highway.

There are many variations of Langton's ant to explore.  even multiple ants interacting.  What might happen if we had 10billion Langton's ants of a dozen different varieties interacting?  hmmm???

go over these laboratory exercises carefully and take time to mull them over.  While none of these chemical or mathematical games comes even REMOTELY close to the molecular complexity of the simplest living cell, they give at least me, a hunch that we are on our way  inventing such games or even finding them under natural chemical conditions, and thus elucidating how chemistry might be able to elaborate all on its own into life.