Canadian bee breeders’ perspectives on growing a domestic queen industry

To best understand how marker-assisted selection (MAS) can be used by the industry, we solicited information from queen breeders about their experience with breeding and selling queens. In early 2016, we developed and launched a queen breeder survey and by early 2017 we had responses from 51 queen breeders, representing approximately 6-12% of our estimated breeder pool. Here is what they told us.

The majority of survey respondents keep their bees in Ontario and British Columbia (Figure 1). They keep a range of different honey bee strains, with the three most common being Carniolan, Buckfast and Italian. The vast majority of survey respondents perform some form of selective breeding, indicating that investment in this process is worthwhile for breeders.

Figure 1. An overview of survey participants, showing their locations (A; brackets indicate absolute numbers of survey participants) and the strains of bees they breed (B).

We wanted to know what qualities breeders actively select in their breeding programs. As you know, one of our goals in the Beeomics project is to expand our marker panel to enable selective breeding for additional traits. Now, we see that the traits we’re developing markers for overlap with the top four that bee breeders are interested in (Figure 2). No doubt some of the other traits we’re working on (grooming, mite resistance, other disease resistance) would be farther up the list if there were currently tools to select for them.

willingness to select
Figure 2. Summary of bee breeders’ interest in selective breeding, including their current experience, desired traits and interest in MAS.

Over two-thirds of the surveyed breeders indicated that they would consider sending their bees for lab testing if it means they can select for these important economic traits. We aim for the cost of this service to be approximately ~$30 per sample, which is around the average price breeders have indicated they’re willing to pay for MAS testing (Figure 3).

Surprisingly, only 22% of the surveyed breeders indicated that they negotiate queen prices with their buyers, with the majority of breeders setting their price ahead of the transaction either independently or based on other local breeders’ prices. Breeders are therefore in a position to set a price that reflects the actual value of their queens. According to 117 surveyed beekeepers (queen buyers), the value they’re willing to pay for a local queen with their top two most desirable traits is $68.

queen prices
Figure 3. The queen breeding business. A) Queen prices amongst survey responders. B) Factors affecting queen price.

One of the barriers to expanding the Canadian queen industry frequently cited by beekeepers and breeders are our prohibitively long and cold winters, necessitating a large number of bee imports to feed our early spring demand. This rigid concept of timing was called into question when beekeepers were asked when they needed their queens (Figure 4). Seventy-five percent of beekeepers need queens between May and June. Surprisingly, when breeders were asked when they could have queens ready for sale, they were largely able to meet the beekeepers’ timeline with 68% of breeders indicating that their queens could be ready for the spring wave. The significant overlap between the supply and demand for Canadian queens directly contradicts the idea that winter and timing is a real barrier to developing our Canadian queen industry further.

queen availability
Figure 4. Queen supply and demand dynamics and other barriers to expanding the breeding industry. A) Bubble size is proportional to queen supply (green) and demand (blue), which largely correlate throughout the season. B) Despite being able to match early queen demand, weather and timing are viewed as significant barriers to expanding the queen breeding industry.

Canada’s agricultural and economic prosperity depends on healthy and productive bee colonies, which in turn depend on healthy and productive queens. Excitingly, this is the first year that queens produced by MAS are on the market. The queens are currently only available in British Columbia, but we hope to expand our services soon and we’re looking forward to using these informative survey results from beekeepers and bee breeders to help us solidify the MAS breeding model, creating a tool that best addresses the industry’s needs. Thank you all for your continued support.

Spring 2017 update

We are approaching our goal of finding gene and protein markers for economically beneficial characteristics. In the last update, protein extraction was about to begin; now, it has been extracted from well over half the antenna samples, which means there are only several weeks left until we can start the analysis. This will let us finally start finding which proteins (markers) correlate with the suite of different traits that were meticulously recorded for each colony last summer. Some people began the important task of optimizing the protein analysis step (which will currently take about 8 months) to cut the time down to a more reasonable length. This is important because when this analysis is offered as a service to beekeepers – our ultimate end-goal – we want the turn-around time to be as fast as possible. The actual time will depend on a combination of demand, optimization and the requirements of queen breeders, but our target is about 2-3 weeks.

Keep an eye out for a new article authored by the BeeOmics team which describes the economic benefits to beekeepers who adopt the marker-assisted selection model. Conveniently, marker-selected hygienic queens are available for purchase this year from four BC queen breeders. If you missed the advertisement you can still find it here, along with the queen breeders’ contact information.

Winter 2016 update

As we have reported previously, we have been very busy tackling the mountain of antenna samples that need to be dissected to analyze their proteins and ultimately find markers for selective breeding. Now, we are happy to report that all the antenna samples are finally processed and we’ve moved on to the only slightly less daunting task: protein extraction. To put the scale of this project into perspective, over 50,000 individual bees have now been dissected, making a total of ~2,700 samples to analyze. This is a mind-boggling number of samples: after the protein is prepared, it will take one mass spectrometry instrument (the piece of equipment that lets us identify and measure the proteins) about 250 days  to complete the analysis. However, the data to come out the other end will certainly be worth it!

The results are in: Support for marker-assisted selection breeding model

Hello everyone! Thank you very much to all the beekeepers who participated in our survey. And if you have not yet participated, it’s not too late (you can still find the survey here)! So far, we have had well over 100 respondents collectively representing 15% of all managed Canadian honey bee colonies. Here is what they had to say.


First, some preliminary information. We wanted to know more about the current state of the industry, namely, how colonies and queens are replaced and where they come from:




The system that is currently in place is very likely going to remain the cheapest option for beekeepers, but we wanted to know if there is a desire for higher queen quality, local queens, and if people will be willing to pay extra for queens satisfying their top two most desired traits. The responses to these questions were very encouraging, showing that there is indeed a demand for the service we are developing!




In order to benefit beekeepers the most, we wanted to know what traits are considered the most important to strive for. The results show that honey production, gentleness, overwintering success and hygienic behaviour are the top four most desirable traits.




To get a sense of what the perceived risks in the industry are, we asked “What are the biggest risks to sustainable beekeeping in Canada?” The top risk factors outlined here overlap partially, but not completely, with the actual causes of colony losses (as self-reported) over the last two years.




Finally, since marker-assisted selection is a new technology to the beekeeping industry, we wanted to know if the concept was supported or not. We are happy to see that despite having very little experience with it, beekeepers are generally interested in sending samples for MAS testing.




The combined results of all these questions tells us that there is a solid foundation of willingness, demand and interest to support kicking this service off the ground. Thanks again to everyone who participated!

Calling all bee breeders!

Queen apis mellifera honey beeDo you breed honey bee queens in Canada and want to contribute to Project BeeOmics? Do you have time for a 15-20 minute survey? If so, we would love to hear from you! The ultimate goal of Project BeeOmics is to offer a new molecular diagnostic service that can guide selective breeding for beneficial traits, including colony survival, temperament, honey production and pathogen resistance. Before the service can be implemented, we need to gather more information from breeders – from hobbyists to full-blown commercial operations – who might want to take advantage of this technology. Even if you only breed queens for your personal use, your feedback here would be much appreciated. This information will be a very important tool to help us understand the current state of queen breeding and what challenges and opportunities lie ahead as we strive to strengthen this industry.

By taking our survey, you are helping to shape the future of this technology. Everything from your timing for queen production, what you consider a reasonable cost for molecular diagnostics, or if you’re even interested in this service at all – these factors will all come into play in getting this service off the ground. Ultimately, this is a tool for breeders, and we need you to tell us how you can make the best use of it.

We are still looking for breeders to participate – if this sounds like something you are interested in, please take our online survey or contact Dr. Miriam Bixby ( if you have any questions.

Q&A with Leonard Foster

ljfOct. 15th, 2016

Leonard Foster, the leader of Project BeeOmics, talked to beekeepers and scientists at this year’s BC Honey Producers Association annual conference. A lively discussion with the audience ensued.

Q: Are you worried about breeding a bee that is too hygienic?

A: Being too hygienic is not likely a problem, but what we do need to be thinking about is that we don’t want to select too strongly in one direction to the detriment of other traits. That’s one reason why we’re expanding the marker panel to include things like honey production and gentleness, so that we can find out if this is happening and hopefully balance it if it does.

Q: Will this kind of selective breeding reduce genetic diversity?

A: Theoretically, but our goal is not to create a single hygienic stock. Instead, we are creating the tools for queen breeders to use themselves – if breeders across the country use the tool, rather than a common stock, then diversity should be maintained. Bees have been selectively bred for many hundreds of years, but domestic honey bees do not have unusually low genetic diversity.

Q: Are projects like this being done in other places?

A: As far as we know, this is the only marker-assisted selection work being done with protein markers. There are some groups interested in similar topics in the States but we try to work with them, rather than compete.

Q: Your hygienic bees had much better overwintering survival even with very high Varroa levels going into winter. Were the surviving colonies thriving or were they just treading water?

A: The surviving colonies were actually quite strong and the Varroa levels had gone back down in the spring.

Q: You are using protein markers for breeding. Have you considered micro RNA too?

A: Not yet, but that is something that’s in progress.

Q: Using diagnostics for selective breeding sounds attractive, but would a “let it be” selective method based on survival achieve similar results?

A: Maybe, but a queen breeder can’t afford to lose that many colonies. Using markers would also let you do more than one round of selection per year and look at many traits at once. Survival is the bottom line but once that’s under control, it would be great to select for other traits as well, which may or may not influence survival (e.g. gentleness, surplus honey).

Fall 2016 project update

Project BeeOmics is marching ahead but has suffered some set-backs over the last few months. It was a very sad day when we received a bee sample shipment from Quebec and found, to our horror, that the samples were useless. Apparently Purolator was overwhelmed with shipments because of the bubbling talk of a strike at Canada Post, and they fell far behind their deadlines. For most shipments, that wouldn’t be a serious problem, but these precious samples that many people worked days or weeks to prepare had to be kept frozen. This is because a major goal of Project BeeOmics is to analyze the proteins in bee tissues, looking for ones that correlate with useful traits; however, the proteins break down at room temperature (think: rot). Of course, sitting in the depot for an extra weekend meant all the dry ice evaporated and the samples became degraded and unusable. Very sad. Also, with the end of the summer came a big loss in helping hands – our summer students have helped immensely with processing the samples, but there were too many samples to complete by the end of the summer. In time, we will get there, but it means there will be another issue or two before we can put together some preliminary reports to share.

Summer 2016 project update

Project BeeOmics is now well under-way, with all the BC colony samples already processed and the Alberta samples rolling in. Manitoba, Ontario and Quebec will be following soon. From these samples (and with a lot of extra helping hands), we are starting to get a look at the molecular differences between the colonies. Once all the data is in, we will relate this information to the other traits we’re measuring (aggression, honey production, innate immunity, hygienic behavior, etc.). What we hope to find are groups of genes or proteins linked to each trait, creating a molecular signature, or fingerprint, that we can use in the same way we are now for hygienic behaviour. When the project is complete, we should be able to take samples from a beekeeper’s breeder colonies, read their molecular fingerprint and use this information to suggest which colonies to breed from to get the nicest, healthiest bees in the next generation. All without taking rigorous notes or doing your own tests! We are excited to share some preliminary data soon.

Spring 2016 project update

Project BeeOmics is kicking off across Canada this summer, so we will have lots of exciting updates to share throughout the coming years. This $7.3 million project will involve ~1,000 colonies all the way from BC to Quebec. Based on our previous success with breeding for hygienic behaviour, the goal of Project BeeOmics is to develop selective breeding tools for 11 additional beneficial traits. Right now, protocols are being finalized and equipment has just been shipped out to the provinces. BC will be leading the pack by testing colonies for aggression, Varroa infestation, hygienic behavior, innate immunity and a suite of other characteristics. With the field season already here (early April swarms again this year!), the UBC team is rushing to get all the supplies ready and start testing!