I've previously presented at PyConAU and related conferences on MicroPython, Internet of Things, Visual Programming, Functional Programming in Javascript, NoSQL in Postgres.

Recently I've been working in bioinformatics. Bioinformatics is the study of biological systems using numerical analysis. Biological systems are pretty complex, so typically this analysis requires computers. I'm been lucky enough to get involved in that part.

Thanks to my colleagues at Walter and Eliza Hall Institute of Medical Research and at University of Washington Genome Sciences and at the Brotman Baty Institute for their patience while I've found my feet in this field.

Mistakes and oversimplifications are all mine!

Bioinformatics is the numerical study of biological systems. Since biological systems tend to be pretty complicated, this involves computers.

Now you've got two problems.

A lot of the foundations of both fields were laid in the 19th century. Gregor Mendel worked out that inheritence worked using discrete inheritable units, he didn't use the word 'gene' but that's the idea. While Charles Babbage and Ada Lovelace were working on the idea that machines could perform calculations, Charles Darwin and Thomas Huxley were working on natural selection and how life could possibly have gotten started.

In the meantime, the DNA molecule got discovered, an enormously long molecule with no known purpose. We'll come back to that shortly.

Nikolai Koltsov proposed a giant molecule as the store of genetic information, but this didn't fit well with the Lysenkoist theories of Soviet Russia, landing him in hot water with his government and leading to his early death.

In the meantime, Alan Turing and Claude Shannon worked out theoretical and practical bases for computing, leading to the invention of cryptanalysis and the cracking of the Nazi Enigma codes. After the war, Turing would turn his attention to morphogenesis, the way organisms grow complex structures, but also ended up getting government attention and an untimely death.

Proteins were known to be composed of a long string of amino acids, and thought to be something to do with inheritance. Frederick Sanger earns a Nobel Prize for working out how to read these sequences.

However, it is soon discovered that proteins are a later step in the process, and the true carrier of genetic information is DNA.

Watson, Crick and Rosalind Franklin work out the structure of DNA and unperturbed, Frederick Sanger works out how to read DNA sequences, earning another Nobel Prize.

In the 2000s, we work out how to edit DNA, eventually culminating in editing DNA in living beings to cure genetic diseases.

The basic unit of life is the cell.

A cell is a bit like a tiny computer.

Inside there's a bunch of programs called genes which tell the cell how to make proteins, which are the working machinery of the cell.

The cell is bounded by a membrane which keeps the insides in and the outsides out, and the cell interacts with the world through channels which are kind of like ports which selectively let specific molecules in and out of the cell.

Bacteria, animals and plant cells are a bit different, but they have these features in common.

They zip together in pairs, A complements T and C complements G, so we generally refer to this smallest piece of genetic information as a "base pair". This is the unit we'll use to compare genome sizes. Because there's four possible pairs each "base pair" is equivalent to 2 binary bits of information. So "1 million base pairs" that means 2 megabits of information.

Unicellular organisms reproduce by copying, and errors arise during the copying process, sometimes leading to novel features in the cells. Substitions, Insertion, Deletion, Duplication and more unusual errors like inversion cause changes in the DNA.

These changes can cause a gene to be shortened or lengthened or split into pieces or fused with another gene. Bigger errors can also result in multiple copies of a gene appearing, and then these multiple copies can evolve in different directions. Changes can lead to parts of the genome effectively being commented out.

On top of this, pesky things like retroviruses can introduce new genes entirely, effectively writing themselves into a cell's genome to get the cell to make more retroviruses.

It's pretty rare that any of these changes are helpful. Most often, they just result in a broken gene. But over vast timescales, the helpful changes add up. Good features thrive, bad features dwindle, and that's natural selection, leading to evolution.

Bacteria also exchange genes with different bacteria or even other organisms, by passing *plasmids*, which are like mini chromosomes. This is called *horizontal gene transfer*.

It may seem counter-intuitive to give away your code to your competitors, but consider: the thing evolution is optimizing here is not the organism, but the gene. If a gene which confers, say, antibiotic resistance gets copied into a new organism and that organism thrives, then the gene is successful even if the old organism is out-competed by the new. The *gene* will get passed on to more organisms.

Yep, Free Software is billions of years old, and there's been a whole lot of cutting and pasting from the primordial script archive.

At some point about 3 billion years ago, a particularly entrepreneurial branch of the Archaea developed a more sophisticated internal cell structure. These are the *Eukaryotes*.

Animals (including humans) are Eukaryotes. Also plants, fungi, algae and slime molds.

The Eukaryotic genome is protected inside a *nucleus*, and specialist *organelles* perform specific functions within the cell. They're linked together by the *cytoskeleton*, a network of filaments and tubes which spans the inside of the cell.

You can think of these as peripheral controllers or coprocessors supporting the main processor, linked by buses. DNA to RNA transcription occurs in the nucleus, and then RNA to protein translation occurs in the *endoplasmic reticulum* surrounding the nucleus.

Yep, we've just added a separation between kernel and userland programs.

Also, some of these organelles, the *mitochondria*, appear to once have been independent Proteobacteria, which were engulfed by the proto-Eukaryotes and instead of being destroyed they were put to work.

Mitochondria are like little containers with their own separate genome. They have their own DNA outside the nuclear DNA. They have their own replication, transcription and translation mechanisms. The cell benefits from their ability to produce ATP, and the mitochondria benefit from the cell's protection.

But they're effectively running their operating system, like little containers within the host cell. They do their own RNA translation, using their own tRNA, and they happen to use a slightly different translation code to the host cell. They've been engulfed but they're still separate and running their own OS.

We like to think of vertebrates as the pinnacle of evolution but plants have done this twice! In addition to harnessing proteobacteria as mitochondria, plants have harnessed photosynthetic cyanobacteria as *chloroplasts*.

You're probably familiar with the phrase "Embrace, Extend, Extinguish"[^eee] referring to the way proprietary systems can use and then destroy open ones. This may be happening here too: there's evidence that the functions of mitochondria are slowly migrating into the nuclear DNA and being lost from mitochondria.