install theme


Signals - Visualizing Cancer Protein Networks

Although there’s (only?) ~25,000 genes in the human genome, our bodies are estimated to make somewhere between 250,000 and one million different proteins (although any given cell probably only expresses maybe 10,000 at a time). This is because almost any given gene can have its coding message (called mRNA) shuffled in several different ways, resulting in different protein sequences from the same original gene.

That’s a lot of proteins. How can we visualize the myriad networks by which these proteins interact and co-regulate the cell and each other? How do you connect a million dots to create a meaningful picture of the cell’s machinery? Animations like this represent efforts to draw that picture.

It shows the interactions between proteins in a cancer cell. Cancer cells are constantly evolving, becoming evolutionary mosaics of broken cells, each competing to beat out its neighbors. Some cancers require a malfunction in only one protein to go awry, but most require multiple errors to occur. In the animation, the connections and interactions themselves change over time as the cancer cell continues to morph and evolve. These manifest as shifting arcs and lines.

(video by Casey REAS)


Big Week for “Synthetic” Biology

A jellyfish made of silicone, and a bacterium made in silico

Synthetic biology is traditionally thought of as repurposing existing or designing new biological parts to do novel things. But in a larger sense, it can be thought of as the ability to create biological systems outside the limitations of pesky things like global and evolutionary time scales. This week marks two really stunning bio accomplishments, each fitting into their own definition of “synthetic”.

Whoa, Jellyman: Cal Tech and Harvard biophysicists announced that they had created a sort of “synthetic jellyfish” this week (pictured above left). By taking thin, carefully designed sheets of silicone and layering rat heart muscle cells over them, they were able to make a bell-shaped living device that pulsed and swam just like the bell of a jellyfish.

Heart muscle cells, or cardiomyocytes, naturally grow together in sheets and will automatically “beat” in a petri dish (with the help of a little calcium). If you provide an outside voltage (like a pacemaker) they will beat in unison! The rat-heart-silicone “medusoid” shape contracted, with the beating cells pulling on the silicone substrate just as a jellyfish’s own muscle cells act on its bell to swim. 

Of course, this isn’t a real jellyfish, but for extra credit you can read Ferris Jabr’s take on what it would actually take to build one.

Byte-size Bio: The other big news this week comes from Stanford and the J. Craig Venter Institute (gracing the cover of Cell this week, above right). Not content with making the world’s first synthetic organism and synthetic genome (Venter’s ambition knows no bounds), they decided to build a computer model of an entire bacterium. Well, mostly.

They modeled, on a very general scale, the tiny bacterium Mycoplasma genitalium, which only has 525 genes compared to our ~20,000, and all of its internal processes on 128 computers operating for 10 hours. To complete a single cell division, it required half a gigabyte of data. But you have to be careful before you call this a completely “simulated organism”. Normal cells have many, perhaps hundreds, of just different types of genes, and they interact in myriad ways … we have just begun to scratch the surface of those networks. Just look at how complicated even the tiny changes in a cancer cell can be!

By simplifying their model down to 28 minimal systems, their computer program matched the bacterium’s biology as we know it. But a more “realistic” model is going to be exponentially more complicated. Here’s some collected reactions at Tree of Life. But, still … wow!

Modern biology has done a very good job at describing the function of individual genes and proteins, but our next chapter lies in how these interactions build into systems. The “-omics” era will be one where we map how the thousands of parts that we are made of combine to make us whole.Simulations like this will be at the leading edge of that era. But we have a long way to go … how many computers would it take to model the trillions of cells in the human body?