A city of 90,000 in southern Germany is an epicentre of innovation in the artificial intelligence space. It’s also the birthplace of a company that is moving the needle when it comes to plant breeders using it.
It takes special people to give birth to a business in the plant breeding and bioinformatics space, but it takes a special city to inspire those people in the first place.
“The city of Tübingen in southwestern Germany is that very place. Located on the banks of the river Neckar in the Stuttgart region, it has always been a unique spot to live and work, enjoying a small community feel while being home to many highly skilled and knowledgeable people in the scientific field”, says Daniel Huson, professor of algorithms in bioinformatics at the University of Tübingen and member of the scientific advisory board of Computomics.
Its history has everything to do with its status as the Silicon Valley of Germany. Over the centuries, as Tübingen grew and developed, it became a center of trade and commerce, as well as an important academic and cultural hub. In the 15th and 16th centuries, the city experienced major growth and prosperity, with the construction of many important buildings and landmarks that can still be seen today.
It’s now home to Cyber Valley, Europe’s largest research consortium in the field of artificial intelligence (AI) with partners from science and industry. Bringing together academic and corporate expertise in artificial intelligence and machine learning, the community-oriented environment in Tübingen allows for both professional and personal relationships with colleagues.
It’s where Computomics was born in 2012, and where the revolution in machine learning for the plant breeding sector was started.
Ahead of the Curve
Let’s rewind to 2002. Tübingen’s academic programs, particularly in bioinformatics, have always been ahead of the curve, and the influx of students due to the biotech boom led to strong student numbers. At that time (and it’s still the case), the city did not have a dominant large company that stood over everyone else. “That lack of a physical and intellectual monopoly fosters an open and innovative atmosphere that encourages entrepreneurship”, says Sebastian Schultheiss, Managing Director of Computomics.
The core group of Computomics founders — numbering six in total, including Schultheiss and Huson — found themselves working together on campus at the University of Tübingen and the Max Planck Campus.
“We recognized that the emerging field of genomics and the vast amounts of data generated by modern sequencing technologies could provide valuable insights into the genetic basis of plant traits, such as yield, disease resistance and nutritional quality,” Schultheiss says. “So, Computomics was formed with the aim of applying cutting-edge computational methods to solve problems in plant breeding and genomics.”
At the time, plant breeders analyzing these massive data sets needed sophisticated computational tools and expertise to do so, and these were lacking in many plant breeding programs.
The Computomics founders had a vision to bridge this gap by providing state-of-the-art bioinformatics and computational biology services to plant breeding programs, helping them to leverage the power of genomics for crop improvement.
“Of course, no company sees success overnight. The challenge of securing funding and acquiring customers while building the business was real,” Schultheiss says.
Computomics was able to grow with the help of stipends and other support to get things off the ground. The company grew by exploring plant breeding as a potential area of focus and establishing a long-term partnership to learn from customers’ needs and improve their product. The team received public funding to support their development and data analysis efforts, which are crucial to bioinformatics due to the large datasets involved.
As the founders worked to build the company, which officially opened its doors in 2012 as a spin-off from the Max Planck Institute for Biology and the University of Tübingen, their technology and expertise attracted major players in the plant breeding and seed industry.
In 2016 Computomics introduced their groundbreaking machine learning technology xSeedScore®, which allows seed companies to optimize traits like yield, flavor, texture and shelf life while offering the ability to improve crops and provide supply chain continuity as crops can be adapted for multiple locations and future climates.
“Tübingen remains the heart of Computomics as a company, and the home of its data center. With the experience of more than 180 successfully completed projects for customers in 18 countries, Computomics helps navigate the complexities of all crops,” Schultheiss says.
Their advanced machine learning methods enable rapid understanding of genomic data for researchers including those in the plant breeding and seed spaces, he adds.
The Rise of Artificial Intelligence
Right now, AI is on a lot of people’s minds as chatbots grab headlines all over the world. It’s important to note that AI and Tübingen have been intimately linked for years. The Cyber Valley initiative was launched in 2016, with the aim of advancing research in AI and machine learning, as well as promoting the transfer of knowledge and technology from academia to industry.
The initiative is based on a collaboration between the University of Tübingen, the Max Planck Institute for Intelligent Systems, and the University of Stuttgart. “Because Tübingen allows for both professional and personal relationships with colleagues, it’s a prime location for AI researchers and companies to establish themselves”, Huson notes.
This has led to the emergence of many innovative AI companies in the area, with Computomics being one of the first AI companies founded a few years before the establishment of the Cyber Valley.
AI systems can be trained on large datasets to learn patterns and make predictions or decisions based on that data. There are different types, including machine learning, deep learning, natural language processing, computer vision, and robotics, and they are being used in a wide range of applications, from virtual assistants and self-driving cars to medical diagnosis, financial forecasting, and in the agricultural space.
When applied to plant breeding, AI can save breeders both time and resources by more precisely identifying the elite genetics to test in the field. AI can capture genetically what would take breeders multiple cycles of selection and large expenditure of resources to achieve.
AI has the capability when applied with discipline and focus to both increase the efficiency and the genetic gain of a breeders pipeline.
The machine learning technology xSeedScore analyzes massive amounts of genomic data, allowing plant breeders to select the best plant traits for desired outcomes such as yield, flavor, texture and shelf life. This technology helps to speed up the plant breeding process by allowing breeders to identify the best genetic combinations for a specific trait and select plants for crossbreeding.
It is a modern use of technology that has been in development in Tübingen for a long time and which will only accelerate as time goes on. Computomics technology is a great example of how AI can be used to benefit all mankind.
—Sebastian Schultheiss and Daniel Huson are pioneers in the field of bioinformatics and co-founders of Computomics
