Some papers seem to be in need of translation – at least in the journal club which I attend. “Programming biomolecular self-assembly pathways” by Yin et al. is one of those, which leads me to write this review. Our debate centered around the question why this paper is Nature-grade, and in my opinion, the answer is contained in the introductory paragraph, which I would paraphrase as:

“Natural self-assembling complexes perform dynamic functions (they follow defined sequences of steps). Engineered self-assembling complexes reach prescribed target structures (without concern for the intermediate structures). Here we program molecular self-assembly pathways.

Let’s talk about the details …”

The point of the first sentence is that natural self-assembly can be completely unlike throwing a bunch of pieces into a pot and letting them stick together if they possess complementary binding sites. Often, great care is taken to select a specific piece prior to an equally sticky piece of a different type. For example, ribosomes and mRNA together act as a mechanism to define the sequence of amino acid addition to a nascent protein.

In the design of artificial self-assembled structures, often the relative energy of possible intermediate states is engineered to smooth the potential energy surface, guide the assembly, and guarantee that the desired final structure is the energy minimum. Yin's et al. advance is to develop a generalized approach, which enables researchers to “program” a sequence of assembly steps for their DNA systems without repeated attention to the physical details of the process. 

Defining the assembly sequence is almost antithetical to the anarchic spirit of self-assembly – uncomfortably close to robotics and resembling classical music rather than jazz. However, it has its advantages. Imagine buying IKEA furniture, where an ingenious arrangement of packaged parts prohibits any assembly step except the first (by IKEA logic), after which one of the parts pops out to reveal a new connection, etc. Beyond building really complicated structures (e.g. Bjoern), a defined assembly sequence permits the repeated execution of a simple sequence of steps in order, which is tremendously useful for machines.

How do Yin et al. achieve this on a molecular scale and what are the first applications? It is all explained in the details. But the pursuit of a general method to define a self-assembly sequence is in my opinion a Nature-grade quest, since it is a surprising challenge to my notion of self-assembly.