A clever idea to try to use AI to make cellular machines has produced something being called xenobots. However, the research is getting overblown in some ways. What does this have to do with Zambonis? Read on.
What are xenobots?
A xenobot is supposed to be a cellular robot made from frog stem cells. The robot part seems kind of on the right track at a primitive level based on the data so far. The main area where things get over-exuberant in my view is the claim that the xenobots are novel organisms.
Indeed, the xenobot papers including a brand new one in the journal PNAS (more below) make the dubious claim that the little spheres made of frog stem cells are an entirely new life form.
They are almost certainly something far simpler: just small clusters of frog cells that can move in unusual ways due to cilia. That ability is indeed kind of different from various cellular spheres that scientists have been making in the lab for years or even decades, but not that different.
So who’s right?
The new PNAS paper out last week claims that the xenobots can reproduce. The paper is entitled, “Kinematic self-replication in reconfigurable organisms.”
I’m surprised about the use of the word “Organisms” in the paper title.
What is the evidence they are new organisms instead of just clusters of cells? After all, these cells are genetically identical to frogs. If I was a reviewer I would not have let that “organisms” part go. Reviewers are usually very tough on such claims in the titles of manuscripts. I would have asked the authors to prove that the cellular clusters are new organisms rather than just balls of cells.
In my previous post expressing some skepticism on xenobots I pondered about what it would mean if they could reproduce. It wasn’t clear that it would mean much, but reproduction is a defining characteristic of organisms. Still, many kinds of spheres of cells have been shown to keep growing and make new spheres and cells can “self-replicate.” We grow such spheres in my own lab here a UC Davis.
A past xenobot paper also had claimed that the cellular clusters could self-heal, which seemed like anthropomorphizing of cells to me. There was also talk of the little cell clusters “walking”. Cells can move around and sometimes I admit to thinking of them “walking” around a tissue culture dish, but overall it just seems like the research is imbuing xenobots with a few too many animal-like qualities.
Xenobots are fancy clusters of frog cells, not organisms
Now thumbing through the new PNAS paper again I’m still scratching my head on the new organism claim.
I also watched a YouTube video from the University of Vermont supposedly showing xenobots replicating. It was a fun video, but kind of underwhelming on the reproduction front.
The xenobots look more like Zambonis on ice rinks collecting piles of ice than reproducing organisms. (as a side note, it’s fun to read about Frank Zamboni.)
The cellular ball spinning movement is cool and maybe they could be kind of like cellular robots.
But as best as I can tell based on the data, xenobots are not entirely new organisms. That claim seems like hype to me.
Keep things simple
There’s a saying in medicine (and no, I’m not a physician but I’ve been doing biomedical research for more than 30 years): if you hear hoofbeats think horses, not zebras. Don’t overcomplicate things. Along those lines, if you see spheres of frog cells (see Figure 1 above from the new paper), I say think “clusters of cells” not some entirely new organism.
From a media item on the new paper comes this xenobot claim as well (emphasis mine):
“People have thought for quite a long time that we’ve worked out all the ways that life can reproduce or replicate. But this is something that’s never been observed before,” said Douglas Blackiston, a scientist at Tufts University who assembled the ‘parent’ xenobots and is also a co-author of the study, published today in Proceedings of the National Academy of Sciences.
Never before? That seems like a stretch.
Actually, as I alluded to earlier, embryonic stem cell or IPS cell-generated cellular spheres like organoids and clusters of brain stem cells and precursors called neurospheres can also make more of themselves. There are probably hundreds of papers on those. That doesn’t make them entirely new organisms.
Admittedly it is interesting and different that the spinning xenobots seem to catalyze collections of scattered frog cells but the new clusters don’t look so much like the “parent” xenobots to me.
For another skeptical look at xenobots and far more detail on the new paper check out this great piece in Ars Technica by John Timmer.
Maybe the next PNAS paper will change my mind on being so skeptical?
On the brighter side, engineering clusters of cells that behave in interesting ways seems worthwhile.
They could be living robots.