Twin Keratin Swimmer

Post your creations here. It can be either a cool genome or an interesting ecosystem. Indicate if you designed it yourself or if it was created through evolution, or both :). File, picture, or video is required.
Post Reply
User avatar
wapcaplet
Posts: 306
Joined: Thu Dec 08, 2016 9:52 pm

Twin Keratin Swimmer

Post by wapcaplet » Wed Feb 14, 2018 10:02 pm

This is my favorite evolution-designed organism so far. It is a keratin-protected phagocyte swimmer with several stages of development, evolved to solve the "Countermeasure" challenge.

Note: You probably need the beta (v.94) of Cell Lab to open the genomes and substrates linked below.

The swimmer has a long and complicated ancestry. It began as a basic phago/flagello swimmer, evolved for the "Macrophages III" challenge. It's not too hard to evolve this kind of simple swimmer with "contaminate cells" plus some time with radiation, but getting one that is fairly efficient and doesn't waste much energy with unnecessary splits or extra modes can take a lot of patience.

I evolved several different swimmers in various attempts to solve "Macrophages IV" (which needs a super efficient swimmer) and "Macrophages V" (which needs a fairly efficient swimmer that goes in circles to avoid being killed by the edge).

For the "Countermeasure" challenge, there is already a simple swimmer living there, along with a devorocyte colony that feeds on them. The intended solution is a protected swimmer that can out-eat the other swimmers, and avoid being eaten by the devorocytes.

I started with my circling swimmer that won "Macrophages V". This swimmer looks fairly simple, though it uses 9 different modes. Several of these modes are almost identical: four of them are "egg" phagocytes (1,11,16,34), three are flagellocytes (10,12,35), and two are feeding/reproductive phagocytes (22,37).
CMK1.png
CMK1.png (11.37 KiB) Viewed 1120 times
CMK1.genome

This swimmer did very well in the challenge with no adjustments. It's so efficient, it outnumbers and starves the original purple swimmers in 15-20 seconds of incubation. This definitely qualifies for the "disrupt the ecosystem" part of the challenge, but without protection from the devorocytes, the swimmer can't quite reach the 420-cell goal.

So I began the laborious process of evolving keratin. I put the swimmer in an experimental plate with devorocytes, and let them mutate. Whenever keratin cells would appear, I would "cell boost" them and their offspring, or boost swimmers that showed keratin sometimes in their genomes, basically to reward the genomes that were on the right track.

This didn't work too well, but with patience I eventually got a swimmer with one (temporary) keratinocyte, which would reproduce once, then lose its phagocyte and die.

They could propagate indefinitely, but because each only had 1 child before dying, their population could never grow. But that's OK: by cloning this genome many times under radiation, some eventually began to reproduce more successfully, until I had one that could make one child, then develop keratin for a short time while making a second child. When the second child is born, the parent loses its phagocyte and dies.
CMK2.png
CMK2.png (12.86 KiB) Viewed 1120 times
CMK2.genome

I kept evolving these against a bombardment of devorocytes until something amazing happened: they became symmetrical, growing keratin on each side, alternating with reproduction.

Their life cycle is remarkably simple, using only 5 modes. Mode 18 is the initial (egg) cell.
  • 18pg: 21pg + 3fl (propulsion)
  • 21pg: 38pg + 18pg (reproduction)
  • 38pg: 21pg + 2kt (protection)
Since modes 21 and 38 generate each other, the reproduction/protection phases repeat indefinitely.
CMK3.png
CMK3.png (13.32 KiB) Viewed 1120 times
CMK3.genome

If they live long enough, they go on to make more keratin on each side. Sometimes they end up with multiple keratin cells on each side, but usually the new keratin just replaces the old keratin. The flagellocyte is never replaced, so it eventually dies of old age, leaving behind the head with keratin still attached; these immobile cells can still make new swimmers if food appears under them, or if other swimmers push them around.

Because the swimmers have a brief unprotected phase in their life cycle, devorocytes can still coexist with them for a surprisingly long time, but just barely. Given enough time, they will probably go extinct.

Here is the "Countermeasure" substrate with the final twin-keratin swimmers. I changed their color to blue so they would look nicer, but otherwise their genome was 100% created by evolution: wap countermeasure twin keratin.substrate
User avatar
Emera
Posts: 179
Joined: Fri Sep 29, 2017 1:58 pm

Re: Twin Keratin Swimmer

Post by Emera » Thu Feb 15, 2018 1:34 am

Impressive! :D
POTATOCYTE
User avatar
Megathosto14
Posts: 487
Joined: Mon Dec 05, 2016 4:14 pm

Re: Twin Keratin Swimmer

Post by Megathosto14 » Thu Feb 15, 2018 10:14 am

Wow, just wow :shock:
You know what, I kinda interested in starting this evolution challenge.
Killing one makes one a sinner.
Killing hundreds makes one a hero.
Killing thousands makes one a king.
Killing millions makes one an emperor.
fungus3
Posts: 112
Joined: Sun Oct 09, 2016 1:15 pm

Re: Twin Keratin Swimmer

Post by fungus3 » Thu Feb 15, 2018 1:58 pm

Megathosto14 wrote:Wow, just wow :shock:
You know what, I kinda interested in starting this evolution challenge.
Good luck. You'll need it if you intend to make a sentient swimmer.
User avatar
wapcaplet
Posts: 306
Joined: Thu Dec 08, 2016 9:52 pm

Re: Twin Keratin Swimmer

Post by wapcaplet » Thu Feb 15, 2018 9:50 pm

Megathosto14 wrote:You know what, I kinda interested in starting this evolution challenge.
Go for it! It can be a lot of fun, and lead to some surprises. Some challenges are quite easy to solve with evolution, but plenty of others are frustratingly hard. For example I still haven't solved "Algae III" with evolution, even though all you need is a photocyte that makes (and keeps) adhesin. I've completed one-third of the challenges using evolution alone. But most of the other two-thirds I expect to be very nearly impossible to solve in this way.

The difficulty is thinking of the environmental conditions that will favor the cells you want. I put some comments about this in another thread; many of the cell types have functions that are obviously favored by evolution. For example if there is light in the substrate, then photocytes will be advantageous. But if there is no light, then photocytes will be of no use, and will not sustain themselves in the genome. If there are unprotected cells that can be eaten, then devorocytes will be useful, and can evolve and sustain themselves. But if all the other cells in the substrate are protected by keratin, then devorocytes are totally useless and will not be sustained by evolution.
fungus3 wrote:Good luck. You'll need it if you intend to make a sentient swimmer.
I think a sentient swimmer would be extremely hard to create through evolution alone. As seen above, it can be difficult and time-consuming to get even one particular cell type to be favored and multiply. Trying to get two cooperating cell types to be favored would be exponentially harder (like a stereocyte + myocyte for a bending smart swimmer), and three cooperating cells more unlikely still (like a stereocyte + two well-balanced flagellocytes for a kite swimmer). Perhaps if you can think of a way for each cell to be useful by itself, you could evolve such an organism incrementally. But what use is a stereocyte with nothing to respond to its signals? What use is a signal-controlled flagellocyte with nothing to provide signals? I don't have an answer yet.
fungus3
Posts: 112
Joined: Sun Oct 09, 2016 1:15 pm

Re: Twin Keratin Swimmer

Post by fungus3 » Sun Feb 18, 2018 6:30 pm

How about forcing mutations to favor sentience?
User avatar
wapcaplet
Posts: 306
Joined: Thu Dec 08, 2016 9:52 pm

Re: Twin Keratin Swimmer

Post by wapcaplet » Thu Feb 22, 2018 3:32 am

How would you propose to do that? There are a potentially infinite number of sentient behaviors--how would one of them be chosen to be favored unless it were actually useful to the organism? It doesn't mean much to say "favor sentience", because sentience is not always favorable.
Post Reply