Wednesday, May 13, 2015

Is Biological Engineering? Not yet.

Alternatively titled: Could we be Josiah Willard Gibbs?

We've been tasked to define biological engineering.

What is biological engineering? That question relies on the answer to a host of related questions. What is chemical engineering, or mechanical engineering? What is engineering? (What is?)

The last question is too philosophical, but the second-to-last one isn't. In particular, what distinguishes engineering from science? There are faculty in Courses 2, 6, 10, and 20 who I'd probably classify as scientists more than engineers, but the line is probably fuzzy. I think a popular place to draw the line, and one that I like, is that engineering is what happens when you start using the word "make" a lot. Engineering is a very principled way of making things.

In 20.309 (yes, I did this whole Course 20 thing backwards, whoops), the instructors frequently made reference to department chair Doug Lauffenberger's vision for biological engineering. He likes to use a cute alliterative mnemonic: Measure -> Model -> Manipulate -> Make. The word "make" appears here too, so his definition of engineering agrees with mine, I suppose. Let's run with this workflow and try to flesh it out more in the context of biological engineering.

How do we make measurements? Using techniques borrowed from pure biologists.
How do we manipulate biological systems? Again, using technologies that emerged from the study of biology.
How do we model and make biological systems? Whoops. Roadblock.

At this point, we hit what I think is the fundamental hole in biological engineering. There isn't as of yet a "right" way to model or make biology.

Let's address models first. Solid theory must precede engineering, since good models can only be built from sound theory.

Anecdotally, this seems to hold up.

The major branches of mechanical engineering have had their theoretical foundations pretty well-set since the 17th, 18th, and 19th centuries, and mechanical engineering is indeed the oldest engineering discipline. Electrical engineering didn't exist until Maxwell.

Theoretical chemistry didn't really exist until Josiah Willard Gibbs, a powerhouse of a scientist, basically invented the field and synthesized a lot of statistical mechanics. It's a bit of a shame that he's only remembered for the Gibbs Free Energy, because it's not too much of an exaggeration to say that the rest of 20.111 is also due to him. But surely enough, within 2 decades of Gibbs' chemical revelation, chemical engineering emerged as a field at the Massachusetts Institute of Technology.

So where's the theoretical biology? Does it really exist? I don't think so yet. Biology is hard, and requires a lot of math. In course 20, we learn a lot of physical and chemical theory in an effort to model biological systems, but things get complicated very quickly, and so we frequently hit walls where all of our models break down and we have to look for new parameters. It's hard to engineer without complete models.

We also need to make things. Again, fabrication techniques are crucial to engineering success. Mechanical engineering couldn't progress beyond a point until reliable manufacturing techniques were developed. Electrical engineering exploded when printed circuit boards were developed. Chemical engineers were lost until we could reliably control a reactor.

Where are the biological fabrication techniques? In some subfields, like BioMEMS, they're excellent and continue to improve. But somehow I think calling BioMEMS, microfluidics and the like biological engineering is cheating a bit. Biologically inspired EE and ME, perhaps. When it comes to actually fabricating things with or inside organisms, we aren't quite there yet. In some respects, engineering biology can be like building a radio inside a television: it could work, but it's also really easy to break the TV, too.

Age plays a role in this: we've only had access to high-quality biological data and technology for a few decades rather than a few centuries, so naturally the "make" and "model" parts of the engineering workflow in biology lag behind chemistry and physics.

Biological engineering is poised to become a bona fide engineering discipline, but I don't think it is yet. There isn't a SPICE or a SolidWorks for biology. There aren't PCBs or CAM for biology (though some people might say we're getting close. EDIT: Turns out there is a Bio-SPICE...spoke too soon.).

I don't think we're real engineers yet. But we're getting there. We're building a new engineering discipline from the ground, we're developing the models and manufacturing tools to turn what is traditionally considered the softest scientist into a hard engineering discipline. And that's really exciting.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.