Melnitchouck's Digital Dreams

Will scientists one day be able to download and replicate any crop seeds they want from a bank of digitalized DNA patterns on a computer, both preserving and transforming the 12,000-year-old agricultural heritage of humanity?

Questions like that make Alex Melnitchouck of Olds College in Alberta cautious, even though he has lived through an era of massive changes that once might have seemed impossible.

As the third generation of a family of agricultural scientists who grew up in Belarus in the Soviet Union, he won the last Soviet Student Olympiad competition in biology. “During the Soviet time, everybody believed (that) in the future  tomorrow will be the same as yesterday.”

The future had some surprises in store.

His award happened in 1991 just two months before the official collapse and breakup of the global superpower, rewriting the world’s maps. Many ex-Soviet citizens “were really confused with what was going on, and I was probably one of those people, so I needed to figure out what exactly I wanted to do and where’s a good place for me.”

He ended up getting a PhD in plant breeding. He remembers sitting in front of his personal computer during the second year of his studies, long before advances such as artificial intelligence became a practical reality.

“I thought, ‘Holy smoke, everything in computer science has already been invented, so the next thing that’s going to happen is the revolution in molecular biology and molecular techniques’… And now after 30 years, I need to say I was wrong because I think eventually it will not be about what is better, molecular biology and molecular techniques or computer science, but the combination of both.

But as a seasoned scientist, he describes digital seed banks as “a nice dream – and I don’t think the science is there yet. I think it would be much easier to produce synthetic foods. Mother Nature is such a complex and sophisticated thing, but never say never. Maybe in 20 or 30 years, it will be possible, but we’re not there yet.”


As part of his work as the chief technology officer at Olds College, Melnitchouck is helping to put into practice the agriculture of the future. The college includes a 2,800-acre Smart Farm, which acts as a living laboratory for the interaction between cutting-edge technology and farming.


It allows industry, researchers and students to test and use everything from computers and sensors to autonomous machinery in the real-world setting of a working livestock and grain farm. Melnitchouck and his colleagues are currently working on creating predictive computer algorithms for precision agriculture.


They aim to use artificial intelligence and machine learning on various types of crop data to predict things such as how much fertilizer needs to be applied to limit waste while maximizing yields, he says. 

The invention of farming about 12,000 years ago transformed humanity, turning small groups of hunter gatherers into large organized communities. It sparked successive civilizations and a widening base of knowledge that eventually gave rise to the scientific advances and digital technology of the modern world.


But therein lies something of a paradox. “You know, agriculture is probably one of the least digitalized industries,” says Melnitchouck, pointing to the relatively easy transition of the entertainment industry from analog to digital media.


People who could only go to a movie theatre or watch broadcast TV to see a film in the 1970s can now enjoy one any time they want on their smart phones via online streaming platforms such as Netflix. 


However, applying the digital revolution to farming is going to be much more difficult, partly because it involves being able to gather, process and then understand an enormous amount of different types of complex information, says Melnitchouck.


“I had an argument a long time ago about agriculture,” he says. “Someone told me, ‘Oh, it’s not rocket science. It’s not rocket science.’ And I blew up and I said, ‘Yeah, it’s not rocket science, it’s way more difficult and unpredictable’.”


There are likely more than 140 factors that influence crop development alone, says Melnitchouck. These include soil moisture, organic matter pH, calcium, magnesium, nitrogen, phosphorus, potassium and micronutrients, along with field topography, slope position and direction, wind speed and direction, precipitation, solar radiation, and cloud cover, to name only a few.


Melnitchouck points to a hyperspectral image of the Smart Farm taken in 2020 by the International Space Station.


Although the human eye sees light primarily in three visible bands perceived as red, green and yellow, the image involved more than 200 bands for each pixel ranging from the blue part of the spectrum to the near infrared, he says.


“And one of the main reasons why we did it was just because spectral analysis of field crops brings very important information about the biochemical processes happening inside plants, and it helps us to explain the interaction between plants, soil, environment, things like that.”

It’s part of the Smart Farm’s HyperLayer data collection concept, he says. Researchers believe that by analyzing different types of information from different sources, predictive algorithms can be created to accurately estimate things ranging from the right yield goal to the risk of certain chronic diseases, he says.

Melnitchouck calls agriculture’s digitalization problem “an equation with multiple variables. It’s not just about these technologies that will be able to detect the crop status or the soil status in the field, it’s also about how that information is integrated with the other pieces of the puzzle.”

But even as researchers begin to use digital technologies such as artificial intelligence to unravel the complexity of agriculture in the hopes of finding new high-tech ways to enhance it, he says it is important to remember that farming will remain very much rooted in a hands-on physical reality.

You can turn things such as money or music into digital ones and zeros for your bank account or your smart phone, but you can’t do that for the food you eat, says Melnitchouck. Farmers still have to “grow physical assets, you still have to utilize land and soil, you have to apply fertilizer, you have to utilize crop protection, and things like that,” he says.

“Unfortunately, there is no way to replace that, and it applies to both regular crop production and plant breeding, so like we just discussed, maybe in the future, there will be a technology that allows us to pick out the DNA pattern specific to a certain (seed) variety, and you just click the button and the system will trace a new seed for you, but right now, it doesn’t work that way.”

Melnitchouck says there will likely be two types of agriculture: classical farming that grows crops using normal seeds and procedures, only much more efficiently and productively; and a sector that manufactures synthetic food.

Instead of getting your protein, fat and carbohydrates from cows or plants growing in a field, you will likely be able to widely enjoy chemically synthesized food within the next couple of decades, he says. 

 “I don’t think it’s the right way to say that one of the two is better, one is worse. I think they will co-exist and there will be a demand for both natural food and synthetic.”

But as someone whose last name in Ukrainian means son of a miller, Melnitchouck says it will always be important to preserve humanity’s agricultural heritage through physical seed banks or vaults that store actual crop seeds, even if – and it’s a big if – scientists eventually figure out how to do it digitally, he says.

He likens the rising wave of new technologies for food production to how most people now prefer digital over paper documents. It doesn’t mean we have to destroy the manuscripts from ancient Egypt or the medieval ages, he says. 

“I think the future doesn’t deny the past.”

Contributor: Doug Ferguson