Computer Model of our Ancestry and the Mechanics of Evolution

skuzie · April 29, 2006

I have read a while back an article about a Yale mathematician who developed a computer simulation to find with high probability how long ago our most recent common ancestor lived. What he discovered was that this MRCA lived just a few thousand years ago, meaning there is a person in history around maybe 3000 years ago that ALL of us alive today are related to, this is quite different result then the proposed 'Mitochondrial Eve' which supposedly lived around 150,000 years ago or the 'Y-chromosomal Adam' who supposedly lived between 60,000 and 90,000 years ago.

So why are these two date frames so different? The reason I believe is that finding out our ancestry from genes is misleading, why? Since we are taking into consideration only one linage of your ancestry, the linage that has our Y chromosome (passed by fathers to sons) or the mitochondrial RNA (passed by mothers to daughters). So at any period in time in the past you have 2 individuals that carry the genetic markers that later you will inherit, but what about the rest of your linage? We know that our ancestry lineage grows by a power of two every generation, that means we are omitting allot of people in the past by looking at our current generic ancestry methods.

Just imagine if you were to draw out our ancestry tree, you have 2 parents, 4 grandparents, 8 grand-grand parents, and so on, thus 2 to the power of (number of generations back). So say a generation takes 25 years (this will probably be less as people in the past had kids earlier thus underestimating our numbers) thus 10 generations back (or around 1750s) you had 1024 ancestor lineages, and 20 generations back (or around 1500s) you had 1,048,576 ancestor lineages! Well you get the idea... just to get my point across in the time of Jesus or around 1AD you would have 1,208,925,819,614,630,000,000,000 ancestors (or 2 to the power of 80) if you would draw out your ancestry tree! That number of course does not represent individual people, but lineages, there are many many duplicated people in that number. So statistically speaking 2000 years ago if each of us had so many different lineages how is it NOT possible that there is a common person in there that we all share in our lineage, this is why I think the date is much more recent as the computer simulation suggests.

I believe that a species gene pool gets inverted - combination and recombination of genes, say every 100 generations or so, meaning that you are related to everyone 100 generations in the past ('everyone' that had kids that 'saturated' the population of today - to put it in a blunt way) and everyone in 100 generations in the future will be related to you (assuming you have kids and your genes 'saturate' the future population). This period of 200 generations is a periodic cycle where series of genes gets introduced and discarted in a species for the above example, thus changing or 'evolving' the species as a whole. If for some reason a species 'cycle' is too long, ie a group in the species does not saturate the whole species, then the isolated group is diverged into a new species as the group is no longer able to breed with the original species.

With my high interest in computer simulations and evolution and my background in computer science I started to create simple computer models for the above scenario, I believe that a computer model/engine that is highly robust and scalable would be able to predict population gene flows with very high accuracy if all the variables are set to mirror the real environment. In the (far?) future as we record our lineages in central databases we will be able to feed this computer model 'real' data instead of generating data what we think is representing the real data, so we will be able to see our species cycle in real form, imagine that! that is of course if we dont end up generically modifing our offspring, then our ancestry will begin to loose its meaning.

What I would like to hear are ideas, suggestions, criticisms, questions regarding this topic. :cake: