Computer science has long looked to biology for inspiration in the design of computer systems and software, and the reverse is also true. This paper explores the connection between the biomolecular processes that occur in a cell and the complexities of computer hardware and software. Specifically, the authors explore the relationship between universal computers such as a Turing machine and living cells. After reviewing Von Neumann’s general-purpose computer, the authors describe the molecular process by which proteins are transcribed from mRNA by ribosomes. On the face of it, this seems like a linear process. The translation from DNA to RNA to protein is referred to as the central dogma of molecular biology and is the guiding principle in most biology textbooks that review this process. It is straightforward to correlate these processes with the parts of a general-purpose computer. For example, the authors equate the chromosomes of a cell with the read-only memory (ROM) of a computer. That is, the DNA that makes up chromosomes provides reliable information that can ultimately be used to make proteins. The tremendous variety of proteins that can be made from the thousands of available genes provides the raw materials for a universal cellular computer. Cellular computers are also inherently parallel.

The connection between cellular processes and Von Neumann computers is a useful one to make, especially in light of attempts in computer science areas such as artificial life to study biology in the computer through complex simulations that generate emergent properties similar to those found in nature. This paper is timely, given the rapid pace at which biology and computation are increasingly being brought together in interesting ways. As we uncover additional complexities in molecular processes at the cellular level, it will be interesting to expand the analogy presented here. For example, the recent discovery and characterization of micro-RNA (miRNA) and long non-coding RNA (lincRNA) that participate in gene regulation opens the door to much more complex computing analogies. The miRNA class of short RNA molecules can bind to protein-coding RNA molecules to influence translation of a peptide. Other non-coding RNAs, such as lincRNAs, can bind to transcription factors, thus influencing the regulation of RNA transcription from DNA. Thousands of these regulatory non-coding RNAs exist in mammalian genomes and some may play a central role in determining biochemical and physiological systems. It will be fun to see if these two seemingly disparate disciplines of biology and computer science can continue to collide for the benefit of all fields of scientific inquiry.