Advancing current breeding strategies with digital plant phenotyping

Phenotyping is a really important part of current breeding strategies, but it is often the bottleneck. Dr Jemma Taylor looks at the role of digital plant phenotyping in speeding up this process

The breeding process is never going to be instantaneous and the burning question now is if we are at the point where the speed of breeding is fast enough to keep up with the speed of change we are seeing in the environment.

We need to develop faster methods: if you could screen a phenotype in as high a throughput as you can with genotype screening, progress would become much faster.

Screening plants

You can do the marker assisted breeding and genetic analysis to understand what’s going on but that’s only half the story. There is still the need to grow the plant up and assess how it behaves in the environment you are subjecting it to, to see if it really has the characteristics you were aiming for. In the end, you still need to screen thousands of plants to fully understand what is happening, and that is where digital phenotyping comes into its own.

There are a few phenotyping platforms capable of doing just that kind of high throughput scale, but they are few and far between and generally difficult to access for those on the ground doing the work, it tends to be more for research facilities. However, at CHAP, we have just this type of equipment available, along with the expertise to use it to its fullest capabilities, available for use by anyone with a viable project idea.

Assessing the phenotype

Different plants need to be assessed in different ways, so we need to select the most appropriate digital phenotyping method for the particular crop. For example, if you are growing a cabbage, your main focus is going to be how big are the leaves, how big is it going to grow, how tough does it get depending on the time of harvest, as well as thinking about pests and diseases. This could be done using cameras, mounted on tractors or drones in the field or on gantries in the glasshouse.

Alternatively, if you are looking at a different crop, say a parsnip, your interest is in the root underground. That can be more difficult to measure, you would have to dig it up and inspect on a regular basis. It would be more challenging to carry out regular imaging of roots that are in soil, in the field. However, CHAP’s Plant Phenotyping and Soil Health Facility at Cranfield University not only has a digital phenotyping gantry for above-ground monitoring, it also has access to root monitoring equipment, which would be useful in this scenario.

Ultimately, it would depend on what aspects are important to that particular grower or breeder.

The challenge for digital phenotyping

A lot of breeding is still done by the naked eye and crop walking, and these rely heavily on the experience of the breeder and is time consuming. We need to find a way to implement a phenotyping algorithm that could rapidly identify a crop’s desired traits. This would provide much more rigorous and repeatable crop consistency.

However, the breeders experience and ‘feel’ for the work, will still be critical as the technology will still miss things and won’t be able to replicate the nuance or see the potential of phenotypes that don’t fit into the programme in question.


To find out more about how CHAP can help to take your crop breeding to another level, see also Digital Phenotyping Lab.

Reach out to the CHAP team to discuss potential project opportunities using our Plant Phenotyping and Soil Health Facility or the Digital Phenotyping Lab: email us using the enquiries form at the bottom of our homepage.