Is Cancer a Good Model of Evolution?

[swamidass (S. Joshua Swamidass)]

Continuing the discussion from Bacteriophages, Budding Yeast, and Behe’s Vindication:

In a recent ENV article (Bacteriophages, Budding Yeast, and Behe’s Vindication | Evolution News), @gauger makes a side comment on cancer and evolution.

@Art

“Note: Cancer is not a good model for asking this question in eukaryotes, because the gain-of-function mutations in cancer do not produce structures or traits that lead to anything beneficial for the organism.”

I don’t buy this for a second. @swamidass?

@cwhenderson

Honestly, this bothered me, as well. The sudden out-of-context mention of cancer mutations seems like an uncalled-for sideswipe directed singly at @swamidass. And although these mutations are not beneficial to the host, this fact is largely irrelevant. These mutations are certainly beneficial for the tumor.

In @Agauger, defense, I think she is giving her honest understanding of the science. I do not take this as a personal dig, but just a friendly version of the same push back she often gets here. The fact she is engaging it here is good, as it is good that Well’s has made it one of this Icons of Evolution. That draws more attention to the original article here:

Cancer and Evolution

Of note, we are coming up on the 1 year anniversary of the is article’s publication. During that time it has received quite a bit of engagement by ID. I count about six articles on ENV plus full membership in the Icons of Evolution. That is quite an achievement for a single article. I think the reason why is because this is not an argument they can neglect. If I am right here on evolution, ID does really have problem, at least as it is currently formulated.

I’ve read all these articles very closely Of note, nothing has needed retraction or revision in light of these exchanges. The argument has extended by studying cancer using a DI metric of functional information: Computing the Functional Information in Cancer. We find out, objectively speaking, that cancer generates massive amounts of functional information. Using the ID metric as is, we compute billions of bits of new information. Fixing the theoretical error in the ID metric, we compute hundreds of bits, far more than should be possible if ID information arguments against evolution are correct.

Of particular relevance, cancer evolutions, at a molecular level, much closer to what we see in mammals than, for example, Lenski’s experiments with bacteria. With all the effort being invested in collecting information about cancer, much of which is freely available on the internet, cancer may be one of the more exciting laboratories for understanding molecular evolution in the coming decade. This is data that cannot be dismissed or ignored.

Of course, we expect the DI to rhetorically dismiss all of this. They do not have a direct way of making sense of this evidence, so they have to dismiss it. I’m just impressed and thankful they see it important enough to keep mentioning. This is important too. Cancer makes sense in light of evolution. It shows that evolution can easy generate thousands and thousands of bits of functional information, rewiring expression, and innovating with new functions that benefit cancer cells.

So, @Agauger (and @pnelson) , thanks for pointing people our way on this again. Happy to answer any questions you have about cancer and evolution.

[T_aquaticus]

@swamidass

Of particular relevance, cancer evolutions, at a molecular level, much closer to what we see in mammals than, for example, Lenski’s experiments with bacteria. With all the effort being invested in collecting information about cancer, much of which is freely available on the internet, cancer may be one of the more exciting laboratories for understanding molecular evolution in the coming decade. This is data that cannot be dismissed or ignored.

I would think that cancer evolution could be modeled in the same way pathogen evolution is. In fact, a significant part of the immune system is devoted to seeking out cancer cells and destroying them. It would also be interesting to reverse engineer evolutionary pathways for malignant cells to better understand how tissue organization and utilizations evolved in the first place.