In apple breeding only around 0.014% of seedlings planted will make it to the final round of evaluations. And even fewer become commercially available cultivars. Researchers at Acadia are bringing together biology and computer science to bump that percentage up and make apple breeding and cultivation more sustainable. 

Why is it that so many apple seedlings are told to make like a tree and leaf? Quite simply, the vast majority of the fruit produced from trees isn’t up to market standards. In other words, no one would reach for them at the farmer’s market, and if they did, they’d be disappointed once they bit into it. 

Sometimes the problem is that the apples are too small, or flavourless. Other times the problem is that the fruit is too susceptible to disease or simply doesn’t store well. 

Another important trait that breeders look for in their fruit crop is appearance. Consumers are, frankly, a little vain. We want a pretty apple that, based on its appearance alone, screams “bite me, I'm juicy and delicious.” 

More formally, “external appearance is the first metric consumers use to assess the quality of fresh fruit, which makes shape, size, and color critical breeding targets,” Acadia researchers explain in their study, published this summer in The Plant Phenome Journal. 

And moving from a tiny seedling to an apple tree full of fruit isn’t a quick process; apple breeders wait between 4-6 years, investing labour and resources to bring all of these trees to fruition. Quite the investment for over 99.9% of your half decade of work to go into the compost pile. 

Apples to apples? Not for these researchers

With the goal of producing higher quality apple trees from seedling to orchard, Canada Research Chair (Tier II) in Agri-Food and Sustainable Agriculture, Dr. Zoë Migicovsky’s (Biology) lab is taking a hard look at apple genetics. In one of the most comprehensive studies on apple shape and size ever completed, the lab is working towards determining which genetic markers can be used to predict market desirable traits in apples. 

To begin these efforts, the researchers analyzed data from Canada’s most diverse research orchard, conveniently located a 10-minute drive from Acadia’s campus at the Agriculture and Agri-Food Canada Kentville Research and Development Centre. 

The scope of the project goes well beyond pure biology; an important element of the project is the computational work that goes into analyzing the thousands of images of apples from the research orchard. That’s where Kylie DeViller, an honours computer science undergraduate student at Acadia comes in. 

As part of her co-op program, Kylie got hands-on experience and was an instrumental part of the research team, taking the first stage of the project from initial analysis to publication. She worked in the Migicovsky Lab from May to December 2024, analyzing images of 5,724 apples. First, she marked each image with dots indicating the top and bottom of each sample, which were the data points she would pull from each picture. Then she ran the numbers to determine whether there was any relationship between the shape and size of the apples. 

What they found is that there was no correlation between apple shape and size. Let us explain why this is exciting. 

Most often, plant biologists find that variation in shape and size is linked. In humans, we see this in the way babies have proportionally much larger heads in comparison to their bodies than adults, but these proportion shift as they grow. Plants also often have size-dependent differences in shape. For example, Dr. Migicovsky’s earlier work found that apple leaves had size-dependent differences in shape.

By determining across a diverse collection of over 500 genetically unique apples that fruit size and shape are not linked, Kylie is setting the foundation for breeding apples with unique shapes without sacrificing commercial fruit standards in size.   

“If we found that the variation in shape was correlated with size it wouldn’t be nearly as promising,” Dr. Migicovsky explains. “It was a really exciting discovery." 

Apple bottom genes

So why does this matter? It means that apple breeders will be able to select for market-desirable shape and size without having to sacrifice one for the other, making it easier to breed model apples. 

In the next phase of the project, Kylie is working on her honours research, co-supervised with Dr. Lydia Bouzar-Benlabiod (Computer Science), where she is linking her measurements of fruit shape and size to whole genome-sequencing data for the apples she measured.

Once Kylie’s work is complete, she’ll know which genetic markers an apple breeder might be able to use to help screen their seedlings for desirable shapes and size. The best part is because this screening is based on genetics, it can be done before the trees produce fruit, saving valuable resources.

“If we can even slightly improve the efficiency of selecting for apple trees with desirable traits, that will be huge,” explains Dr. Migicovsky. 

It will take a while for the research to come to fruition—it can take up to 30 years to develop new apple cultivars. But in a few short decades, you could be snacking on the uniquely shaped fruit of Zoë and Kylie’s labour. 

Get the Acadia experience with Dr. Zoë Migicovsky

In Winter 2026, Dr. Migicovsky will be teaching BIOL 4663/ COMP 4923: Bioinformatics. In this course she will introduce the analysis of biological data using computational tools through examples of applications across biological fields such as human health, agriculture, and conservation.