Algorithms To Live By: How We Will One Day Build A Digital Embryon

Algorithms To Live By: How We Will One Day Build A Digital Embryon

by Patrick Landon Ferree, PhD Cell Biology

All animals (including humans) start life as a single cell. It is then the goal of the developmental biologist to work out how exactly that single cell becomes the trillions of cells in the animal body, and how those trillions of cells organise themselves into complex arrangements like organs and organ systems some of which pump blood (like the heart), while others theorise about their own existence (like the brain). What is extraordinary about this process is that it happens reproducibly from embryo to embryo. That single primordial egg cell contains within it the information and means to make arms and legs and eyes and ears, and it does so just about every time.

Knowing what we know today about computers, it is hard to look at animal development and not think about computation and algorithms. It is just too tempting to imagine that the egg follows in some sense a series of step-by-step instructions that guide it to its final form. The instructions might look something like this:

Step 1: divide into two cells.

Step 2: divide into four cells.

Step 738889: grow an arm here and here.

It is an interesting piece of history that we have the same person to thank for both the concept of computation and the concept of mathematical embryonic development. Everyone knows (or should know) who Alan Turing is. He is considered by many to be the father of modern-day computer science and artificial intelligence. He dreamed up the general-purpose computing device called the Turing machine, and he is famous for worrying about how we’d ever distinguish between people and intelligent machines imitating people.

But in addition to all of that, Turing was fascinated by biology and wished to find an algorithmic basis for its behaviors. Back in 1952 (which is prior even to the discovery of the structure of the gene) he wrote down some equations that seemed to show how certain special chemical reactions can start to form patterns that look a lot like the stripes on tigers and the patchy blobs of color on fish. From these observations and derivations, he suggested that similar strategies might one day allow scientists to explain the whole process from egg to adult in terms of mathematical principles.

That’s the dream anyways.

Today we know tons about genomes and their sequences, and proteins and their functions. We’ve characterised and categorised cell types and their interactions with one another. But the holistic picture is still murky, and the story of development is still mostly just a story rather than an algorithm. However, if you squint your eyes and look at the trends in the field, you can catch a glimpse of the future to come. Here are some of the strategies that are taking us there.

First and foremost are the radical breakthroughs in the tools used to visualise the embryonic development of living animals. For the longest time it was only possible to take photographs of dead biological tissues and extrapolate out about the cellular behaviors and their signaling pathways. Now there are new microscopes that allow scientists to literally film limbs growing and hearts beating at the cellular and molecular level. Most of these experiments are carried out in so-called model organisms like nematodes, frogs, flies, zebrafish, and mice. You have to start somewhere and it’s generally a good idea to start with the simplest models first. That’s why so much of what we know about the genetics of animal development comes from research on fruit flies. They grow and reproduce fast (from egg to maggot in 24 hours).

Second are the genetic engineering technologies that continue to hit the market. CRISPR is at the top of the list because it allows scientists to make directed changes to an animal’s genome. This is extremely helpful for studying development because you can modify a gene and then film the animal growing and see how that gene-modification affected the whole process.

Third are the single-cell sequencing technologies that allow scientists to isolate cells in animal bodies and characterise their levels of gene expression. Although all the cells in a body have the same genome (they all have the same DNA molecules), only a subset of those genes are activated (“expressed”) at a given time. In recent years we have been able to measure gene activation at the single cell level across the whole of an organism.

Fourth are the awesome collaborations that exist between biologists, physicists, engineers, and computer scientists. Interdisciplinary research seems to be a staple of our time and with any luck it will bring us a fuller and deeper understanding of biology. It is becoming more and more common for biomedical labs to have scientists with all kinds of backgrounds working on the same projects. Engineers building microscopes. Biologists editing genomes. Physicists building mathematical models. Computer scientists writing software. In an earlier blog post we wrote about the concept of the “digital twin”. No doubt it is a mission of developmental biologists to one day build a digital twin of an embryo.

Contact our experts to learn how we can help your organisation