A miracle child and an Aussie Nobel Prize winner. The five biggest science discoveries of 2025

The chemical’s signal is relatively low and requires some level of interpretation. But more importantly, just because we don’t know a way to produce this chemical, for example through geological or atmospheric processes, doesn’t mean it can’t happen.

This discovery is highly controversial, but Duffy says it has prompted many people to think about the conditions for alien life and how to make sure we’re seeing something that actually originates from life, rather than other, less sensational causes.

AI-made antibodies – Professor Merlin Crossley, molecular biologist and vice-chancellor for academic quality at the University of NSW.

In 2025, biotechnology companies have greatly increased their efforts to incorporate artificial intelligence models into science.

“It’s unclear whether all the strategies will work, but careful data sets in protein databases led to AlphaFold,” says Crossley.

AlphaFold is an artificial intelligence model created by Google and designed to solve a central and very challenging biological problem.

Proteins are the building blocks of our body; their functions depend on their complex shapes. We can read the 20-letter code of amino acids that make them up, but it is very difficult to know their structure without expensive and difficult experiments.

AlphaFold is transformationally good at guessing the structure from code alone; The designers of the model won the 2024 Nobel Prize. “This is garbage in, not garbage out,” says Crossley. “It’s diamonds in, diamonds out. It’s now possible not only to predict how proteins fold, but to start designing them.”

Antibodies are proteins that rely on their unique shape to help bind to and destroy viruses and bacteria. This year, US-based biotechnology company Absci gave the first patients an AI-designed antibody targeting inflammatory bowel disease.

“It’s early days, but the promise of designer proteins is huge,” says Crossley. “This is not some vague AI hype.”

Baby KJ, miracle child

Baby KJ was born with mutations that disrupted the function of a key enzyme; This meant that toxic products slowly accumulated in his body.

“The baby’s future was grim,” says Crossley.

A liver transplant could have cured the disease, but instead KJ from Pennsylvania underwent genetic surgery, becoming the first person to receive such treatment.

Within six months of KJ’s birth, a team of scientists produced a custom gene editor and packaged it into lipid nanoparticles. The editor known as CRISPR Instructions given to scan KJ’s genome until they found the exact DNA that needed to be edited.

“It seems to have worked,” says Crossley. One Study published in MayUS researchers behind the treatment said KJ’s condition was improving and there were no serious adverse effects.

“Millions of people suffer from genetic diseases, and some people whose family members have had their genome sequenced are discovering that they will inherit late-onset conditions. Universal treatments are not yet available, but gene-editing technology and RNA delivery strategies continue to improve. Baby KJ is just the beginning.”

Quantum Cryptography Dump – Professor Nalini Joshi, Head of Applied Mathematics, University of Sydney

In May, Google Quantum researcher Craig Gidney uploaded a preprint article whose title was difficult to parse: How to factor 2048-bit RSA integers containing under a million noisy qubits?

Joshi says this sent a shockwave through the crypto space. RSA is the most widely used encryption algorithm for everything from your smartphone to government secrets; It is based on prime numbers that are 2048 bits long. At this size, it is essentially impossible to break traditional computers.

Quantum computers could theoretically do this much faster. Gidney’s original prediction in 2019 was that you might need 20 million qubits (the key measure of quantum computing power) to break encryption. In his 2025 article he reduced that estimate to just one million.

“This paper shows that the looming threat of widely used cybersecurity protocols being broken down by quantum computers is closer than we think,” says Joshi. “Many have suggested that currently used quantum computers capable of factoring large numbers would need to be so large that they might well be science fiction fantasies and could never be built in our lifetimes. Gidney suggested that this would require only a million ‘noisy’ qubits, the building blocks currently being built by many companies.”

The rise of the dark metabolome – Oliver Jones, professor of chemistry at RMIT

Just as genomics is the study of all the genes in an organism, metabolomics looks at all the small biological molecules (metabolites) in our cells, tissues, and organs.

Loading

“We try to measure as many metabolites as possible and then look at how that metabolic fingerprint changes in response to things like disease,” Jones said. “The goal is to find small changes in metabolism that will allow us to predict potential harm before it happens.”

However, accurate predictions rely on the identification of all detectable compounds, and as analytical methods become more complex, the number of incompletely identified metabolites increases.

This is the dark metabolome, and one guess is that approximately 85 percent of all metabolites it remains uncharacterized with an unknown function in the human body. What secrets do they keep?

“There are too many discussion in community about how comprehensive it is. dark metabolome – and even how exactly we define it. Solving this problem could lead to advances in understanding our biology and that of other species, as well as greater understanding of how biological systems are affected by things like climate change and, my personal interest, the low concentrations of pollutants like microplastics and PFAS we find in the environment. I think this is really exciting; “I’m really looking forward to seeing where it goes next.”