NB: From tomorrow, all non-Monday posts will be set to subscribers-only.

I’ve never been much for video games. My hand-eye co-ordination is terrible; a series of MRI scans in my 40s revealed that the area of my brain responsible for these tasks is abnormally small, which explains why I’ve always struggled with ball games and, indeed, learning to drive. But video games are a huge industry, and many of us consume aspects of them without even playing a game.

The software engines underlying many games – the systems which handle animation, including aspects such as detecting when moving objects overlap, are important tools used by the entertainment industry when making animated films and TV programs. They’re also used to make training videos for the military. And a paper which appeared online last week gives an intriguing new use for them: depicting the structures of large, complex molecules like proteins (https://www.biorxiv.org/content/10.1101/2023.12.08.570879v1.full).

The function of proteins depends to a great extent on their 3D structure. A simple chain diagram, which depicts the chemical formula, shows how the molecule would look if it were stretched out straight, but that flexible chain will fold and coil in on itself influenced by electrical charges on the various chemical functional groups attached to the chain. This creates three-dimensional active sites on the protein, into which other chemicals fit, allowing the protein to carry out chemical reactions. This is the key to metabolism and other processes occuring in a living system.

Predicting folded protein structures was one of the first major successes of AI: the Google subsidiary DeepMind produced an AI programme called AlphaFold which, in 2018, was placed first in a competition to predict folded structures. AlphaFold2 won the contest again in 2020, scoring above 90 for two-thirds of the proteins it worked on, compared with their structures determined by experiment.

Even with this advance, however, it’s still difficult to visualise complex folded protein structures, especially in the environment where they work: inside living cells. In a preprint paper – that is, it has not yet cleared peer review – Muyuan Chen of Stanford University described using Unreal Engine, one of the most widely-used videogame engines, to assist with this task.

One challenge was the resolution needed to depict the structures, Chen explains. The task needs many times the resolution needed to convey a human body moving in three dimensions accurately, he says: this is expressed in triangles, with the software constructing an apparent 3D surface from thousands of tiny flat triangles. A human body requires 100,000 triangles, but “a small scene of a few proteins can contain tens of millions of triangles,” Chen says. A feature of the newest version of Unreal Engine helped with this, he explains. Called Nanites, this feature renders objects in different resolutions depending on their distance from the viewer. Watching the videos Chen has embedded in his paper feels similar to the sequences in everyone’s favourite biological sci-fi film, Fantastic Voyage, but without Raquel Welch in a wetsuit.

The system isn’t perfect, Chen warns. One weakness is that currently it can’t depict accurately the interaction between proteins and lipids, the fatty compounds which form cell walls, and he hasn’t yet automated the placement of membranes or the shape of the organelles which perform the functions of cells. The action of water in the cells isn’t taken into account. The sheer size of the animation files is also an issue: no standalone VR headset can handle them, so they have to be connected to an external rendering computer via a cable. But such systems have promise for researchers developing drugs which work by affecting the active sites of proteins inside cells, for example.

Few things have such a powerful effect on the mind as live music. It can soothe, it can excite, it seems to have a direct hotline to the emotions. And it does so more powerfully than recorded music, even when the performance reproduces a recorded version with high fidelity. The effect is seen across all genres of music, but why this is the case has always been a mystery. Researchers from the University of Zurich believe they have uncovered an important insight into why this happens, and explain this is a paper in the Proceedings of the National Academy of Sciences (https://www.pnas.org/doi/full/10.1073/pnas.2316306121).

Psychologist Sascha Frühholz and his colleagues began by writing 12 short pieces of music, each 30 seconds long, designed to convey specific emotions or emotional states, like calmness, excitement or anger. The experiment’s 27 subjects listened to these pieces, performed live by a pianist or recoded, while lying in an MRI scanner monitoring the activity of the amygdala portion of their brain, responsible for emotions. The pianist could see the feedback of the MRI and could adjust their playing in response.

The researchers found that the live versions of the pieces had a far stronger effect, whether the emotions the music was written to elicit were positive or negative. This was despite the subjects being in the distinctly unpleasant an unrelaxing environment of an MRI scanner. In a real live music environment, musicians don’t have access to MRI scans and just play, but from our own experience we know that witnessing music being played is a powerful experience: the social aspect of hearing music along with many others is also a factor, which the Zurich team couldn’t replicate. A Pink Floyd song describes the experience of attending a rock concert as “the warm thrill of confusion, that space cadet glow”, and while being inside an MRI scanner might well be similar to being in a spacecraft capsule, it certainly isn’t pleasant, and confusion is the last thing an astronaut candidate needs.

Humans are mostly not human. That’s to day that human bodies are colonies of billions of living cells, and the majority of them are of species other than Homo sapiens: many species of microorganism find their homes inside us. We’re becoming more used to this idea of the microbiome, and how it can effect our health, mostly with how it concerns our overlapping digestive and immune systems.

But the microbiome is everywhere, not just in our intestines. Even the brain is home to non-human microorganisms, despite a very efficient barrier between brain tissues and the blood vessels which supply them, and it seems that these microorganisms might be connected to brain disorders like Alzheimer’s disease.

Infective medicines specialist Richard Lathe of the university of Edinburgh’s medical schooled research into the brain microbiome, and has found that it comprises some 100,000 species – about a fifth as many as live in the gut. Based on studies of post-mortem brains held in UK brain-banks, Lathe describes in a pre-print paper that Alzheimer’s affected brains contained unusually high levels of Steptococcus and Staphylococcus bacteria, three types of fungal cells, and an uncharacterised type of alga (plant cell). The bacteria and fungal species were all related to known human pathogens (https://www.biorxiv.org/content/10.1101/2023.02.06.527297v1).

Lathe used a technique he calls the “electronic tree of life”, which matches the result of tests for small fragments of genetic data against databases of microbiome species (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9773549/). Tests of insect, octopus and other vertebrate brains also indicated the presence of microbiomes. He cautions that his results don’t necessary indicate that the microbiome he found caused Alzheimer’s: in fact, although the microbes tended to be associated with the tangles of proteins which characterise the disease, this association “fell short of statistical significance”. Alzheimers might not in fact be a single disease, he adds. Lathe concludes that there might be such things as healthy and unhealthy brain microbiomes. He believes that this might be a fruitful avenue for future research into brain health. To me, it shows that we, and other animals, are stranger than we ever thought. It might even be that other animals literally affect how we think. Raquel Welsh (and her wetsuit) never had to contend with that.

Home

Folding games, wetsuits and the warm thrill of confusion