2. November 2026
Can sperm sort themselves by sex? This is the focus of the new collaborative project “Sexing through Self-Propulsion”. Using experiments, theoretical physics, and computer simulations, researchers aim to explore a new approach to animal breeding – with potential benefits for animal welfare.
X and Y sperm may differ not only in their genetic material, but also in how they move. This is precisely where the new collaborative project “Sexing through Self-Propulsion” at the Jülich Institute for Advanced Simulation comes in. Together with partners from research and industry, the participants aim to find out whether small differences in sperm behaviour can be specifically amplified – so that sperm sort themselves in tiny channels. Prof. Jens Elgeti from the Theoretical Physics of Living Matter Division explains the role computer simulations play in this and how, in the long term, this could lead to a method for animal breeding.
In this project, we are taking basic research from the field of information science – in our case, computer simulations of the generic properties of microswimmers – and applying it to practical use. When I started simulating sperm 20 years ago, I would never have dreamt that we could one day contribute to preventing animal suffering.
Just how relevant this is in practice is evident from the fact that half of the project is funded and carried out by industry partners: djangsters, a software company from Cologne that develops web-based platforms for managing, analysing, and visualising data, and Bayern Genetik, an animal breeding and insemination company from southern Germany that specialises in Fleckvieh cattle and pigs. The project is funded by the Federal Ministry of Agriculture, Food and Regional identity.
In agriculture, one sex of farm animals is often more economically viable. For example, only female cows produce milk; bull calves born to dairy cows are not needed. This means that the other sex has to endure a great deal of animal suffering. Just search for ‘calf problem’ or ‘piglet castration’. It’s not a pretty sight.
The sex of the animals is determined by the sperm: Sperm with an X chromosome produce female animals, whilst sperm with a Y chromosome produce male animals. Almost all farm animals are artificially inseminated, which means that sperm is collected, stored, and dispatched for use. If I sort out all sperm with a Y chromosome before insemination, I will only get female animals. This means, for example, that I no longer need to castrate male piglets.
With cattle, this can already be done using DNA dyes, but this has its drawbacks. The sorting is carried out individually for each sperm cell and is therefore slow. Inspired by the natural fertilisation process, we rely on the sperm’s own movement. We are, so to speak, creating an environment in which X sperm swim in one direction and Y sperm in the other. They sort themselves, in other words. This is faster, more natural, and has further advantages. In particular, it might even work for pigs.
Such a sorting method could therefore enable a wider application of sexing. The fundamental benefits are well known, even in areas where there are as yet no practical applications, such as with pigs. However, much remains to be clarified – we are only just at the beginning. In theory, our approach is promising; in practice, however, there are still many hurdles ahead of us. To bring about a tangible improvement in animal welfare, we ultimately need a scalable application of the technology.
Of course, such a development can only work with real sperm. I would like to express my sincere thanks to Bayern Genetik and Pavlik Lettinga from IBI-4 for their support and the fantastic experiments we are conducting together. We record the movement patterns of many sperm under different conditions using a microscope. The resulting videos and measurement data are collated on the data platform, where they are analysed by djangsters.
Based on this data and our many years’ experience in simulating sperm dynamics, we are developing a digital twin of X and Y sperm. This enables us to test ideas quickly, understand phenomena, and make predictions. The theory and the simulation show us how small differences in swimming behaviour can be amplified in a given environment. For example, we already have simulations showing how sperm with different beat frequencies can be separated in a shear flow.
This means we can test on the computer which structures and conditions steer X and Y sperm in different directions before we build them in the laboratory.
Yes, unfortunately we’re still very much at the beginning. There is data showing that X and Y sperm swim differently – but unfortunately there are also studies claiming the opposite. I think this is due to differing experimental conditions. However, it’s actually not that easy to measure this accurately. I do think, however, that we’re on the right track.
First, we need to develop an analytical setup that allows us to distinguish between X and Y sperm whilst simultaneously tracking their movement. That’s exactly what happens in the data platform. It tracks every single sperm in the microscopy videos – its trajectory, its orientation, and its beating pattern. At the same time, it uses a DNA dye to determine whether it’s X or Y. The difference in DNA content is only a few %, which poses a real challenge in terms of Measuring Technology.
Together, we need to identify the conditions under which differences in swimming behaviour become apparent, and then optimise these conditions. Based on our simulation results so far, I am very confident that the separation will be relatively straightforward once we have identified these conditions. I already have several ideas about how this might work, for example if the beating frequency differs.
In three years’ time, we aim to demonstrate experimentally that separation is, in principle, possible. However, there is still a long way to go before this can be put into practice.
Video description: Sperm can, for example, be sorted in microfluidic channels. In simulations, we were able to show that, for instance, zigzag-shaped channels amplify differences in the beating pattern and that sperm detach differently at corners or curves. Such simulations can help to design structures that can then be realised in experiments.
Further information
djangsters: https://www.djangsters.de/de/
Bayern Genetik: https://www.bayern-genetik.de/de/index.html
