Genetic parameters and genotype-by-diet interactions forgrowth traits in Australian black soldier fly larvae: Implicationsfor selective breeding in the circular bioeconomy

This study demonstrates that while Black Soldier Fly larvae growth traits exhibit low to moderate heritability and significant dominance effects, selective breeding can achieve substantial genetic gain provided that genotype-by-diet interactions are strategically managed.

The Gist

Imagine a tiny, super-efficient recycling machine that turns kitchen scraps and food waste into high-quality protein for animal feed. That's the Black Soldier Fly Larva (BSFL). They are the stars of the "circular bioeconomy," a system where nothing goes to waste.

However, just like any farm animal (cows, chickens, or fish), not all fly larvae are created equal. Some grow bigger, faster, and fatter than others. The big question for farmers is: Is this difference because of their food, or is it in their DNA? And if we want to breed the "super larvae," how do we do it?

This paper is like a genetic detective story that tries to solve that mystery for Australian Black Soldier Flies. Here is the breakdown in simple terms:

1. The Setup: A Genetic "Melting Pot"

The researchers didn't just pick one family of flies. They went to 10 different locations across Australia and collected wild larvae. Think of this as gathering the best athletes from 10 different towns and mixing them all together in one giant gym. They let them breed for a few generations to create a massive, diverse "genetic soup" to ensure they had a wide variety of traits to work with.

2. The Experiment: The "Three-Menu" Test

To see how these flies react to different foods, the researchers fed the larvae three very different diets:

• Soy Okara: The leftover pulp from making soy milk (high protein).
• Brewer's Spent Grain: The leftover grain from making beer.
• Fruit & Veggie Waste: Crushed-up market scraps.

They took 2,097 larvae, fed them these different menus, and measured how big they got (weight, length, width, and surface area). They also took a tiny DNA sample from each one to map out their family tree using a custom "6K SNP panel" (think of this as a genetic barcode that tells you exactly who is related to whom, even without knowing their parents).

3. The Big Findings

A. Food Matters Most (The "Chef" Effect)

The biggest factor determining how big a larva got was what it ate, not its genes.

• Analogy: Imagine two identical twins. If you feed one a steak and the other a bowl of plain oatmeal, the steak-eater will grow bigger. In this study, the "Soy" diet produced the biggest larvae, while the "Fruit/Veggie" diet produced the smallest.
• Takeaway: You can't just rely on breeding; you have to feed them well.

B. The "Hidden" Genetics (The "Seed" Potential)

Even though food was the main driver, the researchers found that genes still mattered.

• Heritability: This is a fancy word for "how much of the size difference is passed down from parents to kids." The study found the heritability was low to moderate.
• Analogy: Think of it like planting seeds. Some seeds are genetically programmed to grow into giant pumpkins, but if you plant them in poor soil (bad diet), they won't reach their full potential. The study says there is definitely "good seed" out there, but it's hidden under the influence of the diet.
• Good News: Because the flies reproduce so fast (a new generation every 6 weeks), farmers can still make steady progress by selecting the biggest larvae to be parents for the next round.

C. The "Teamwork" Effect (Dominance)

The study found something interesting called dominance effects.

• Analogy: Imagine a sports team. Sometimes, two average players (parents) can produce a superstar child because their specific genetic traits "click" together perfectly. This isn't just about inheriting "good genes" from one parent; it's about the combination of genes creating a "hybrid vigor" (heterosis).
• Takeaway: Instead of just picking the best individual, farmers might get better results by crossing different family lines to create these "super combos."

D. The "One Size Does Not Fit All" Problem (G x D Interaction)

This is the most crucial part of the paper. The researchers found Genotype-by-Diet interactions.

• The Metaphor: Imagine a runner who is amazing on a track (Diet A) but terrible on a muddy trail (Diet B). Another runner might be the opposite.
• The Result: A larva that grows huge on the Soy diet might not grow very well on the Fruit waste diet.
• Why it matters: If a farmer wants to use fruit waste (which is cheap and available), they can't just breed larvae based on how they perform on Soy. They need to breed specifically for the fruit waste diet, or accept that their "super larvae" might flop when the menu changes.

4. What Does This Mean for the Future?

• Breeding is Possible: We can definitely breed better Black Soldier Flies to grow faster and bigger.
• Context is King: You can't have a "one-size-fits-all" breeding program. If you feed your flies beer grain, you need a specific breeding line for that. If you feed them fruit scraps, you need a different one.
• The "Super Team" Strategy: Because of the "dominance" effect, mixing different genetic lines (crossbreeding) might be the secret sauce to getting the biggest, healthiest larvae.

The Bottom Line

This paper is a roadmap for turning Black Soldier Flies from a "wild insect" into a high-performance livestock. It tells farmers: "Yes, you can breed them to be better, but you have to be smart about what you feed them and which families you mix together. The food changes the game, and the genes have to match the menu."

Technical Summary

1. Problem Statement

The Black Soldier Fly (BSF, Hermetia illucens) is a critical species for the circular bioeconomy, capable of converting organic waste into high-value protein for animal feed. While selective breeding has successfully improved traits in other livestock and insects, its application to BSF is constrained by a lack of accurate genetic parameter estimates, particularly for the distinct Australasian genetic lineage.

Key knowledge gaps identified include:

• Sparse Genetic Data: Existing studies rely on pedigree-based designs (full-sib/half-sib) in separate rearing trays, which often inflate heritability estimates due to unseparated common environmental effects.
• Limited Genomic Integration: Few studies utilize genome-wide SNP data to estimate parameters in large, mixed populations.
• Genotype-by-Diet (G × D) Interactions: There is insufficient understanding of how different BSF genotypes respond to varying organic waste substrates (diets), which is crucial for optimizing breeding programs for specific production environments.

2. Methodology

Experimental Design & Population:

• Base Population: Established from wild larvae collected from 10 geographically distinct locations across Australia (Nov 2023–Feb 2024) to capture broad genetic diversity. These were pooled and outbred for three generations in a commercial facility to create a genetically mixed base population.
• Sample Size: 2,097 fifth-instar larvae were phenotyped and genotyped.
• Dietary Treatments: Larvae were reared on three distinct diets representing common organic waste streams:
  1. Soy Okara (SYK)
  2. Brewer's Spent Grain (BSG)
  3. Fruit and Vegetable Waste (FVW)
• Phenotyping: Measured traits included Larval Body Weight (LBW), Length (LL), Width (LW), and Surface Area (LSA). Larval color and sampling day were recorded as covariates.
• Genotyping: DNA was extracted from head tissue. Samples were genotyped using a custom 6K Allegro SNP panel designed specifically for BSF. Quality control resulted in a final dataset of 5,009 SNPs for 2,092 individuals.

Statistical Analysis:

• Software: ASReml-R v4 and SAS 9.4.
• Models:
  • Additive Model: Estimated additive genetic variance ($\sigma^2_A$) and heritability ($h^2$).
  • Additive-Dominance Model: Included dominance genetic effects ($\sigma^2_D$) to account for non-additive variation.
  • Multivariate Model: Used to estimate Genotype-by-Diet (G × D) interactions by treating the same trait measured in different diets as distinct traits.
• Relationship Matrices: Genomic Relationship Matrix (GRM) and Dominance Relationship Matrix (DRM) were constructed from SNP data, avoiding pedigree assumptions.

3. Key Contributions

• First Genomic Estimates for Australasian Lineage: Provides the first comprehensive estimates of genetic parameters for the genetically distinct Australian BSF population using genome-wide SNP data.
• Superior Experimental Design: Utilized a "common garden" approach where larvae from mixed families were reared together, minimizing common environmental effects that plague pedigree-based studies.
• Quantification of G × D: Systematically quantified genotype-by-diet interactions across three commercially relevant waste substrates.
• Inclusion of Dominance: Demonstrated the necessity of modeling dominance effects in BSF breeding, a factor often overlooked in insect genetics.

4. Key Results

Non-Genetic Factors:

• Diet: The primary driver of phenotypic variation. SYK produced the largest larvae, followed by BSG, with FVW yielding the smallest.
• Sampling Day & Color: Significant effects were observed; larvae lose weight and darken after the fifth instar, indicating that harvest timing and larval color are critical management variables.

Genetic Parameters:

• Heritability ($h^2$): Estimates were generally low to moderate.
  • Additive-only model: $h^2$ ranged from 0.08 to 0.21.
  • Additive-Dominance model: $h^2$ decreased to 0.05–0.14 overall, but diet-specific estimates were higher (up to 0.36 for LBW on SYK).
• Dominance Effects: Significant across all traits (dominance ratios 0.09–0.19). The inclusion of dominance reduced additive variance estimates, suggesting non-additive genetic variation is substantial.
• Genetic Correlations: High and positive among growth traits (e.g., LBW and LSA correlation = 0.98), except between Width and Length (0.51). This implies that selecting for body weight will effectively improve surface area.

Genotype-by-Diet (G × D) Interactions:

• G × D interactions were moderate (genetic correlations between diets ranged from -0.04 to 0.49).
• The strongest correlation was between SYK and BSG ($r_g = 0.43$ for LBW).
• Correlations involving FVW were weak and often non-significant, indicating that genotypes performing well on high-quality protein diets (SYK/BSG) do not necessarily perform well on low-quality, high-fiber diets (FVW).

5. Significance and Implications

For Breeding Programs:

• Feasibility: Despite low heritability, the short generation interval of BSF (38–45 days) makes long-term genetic gain feasible.
• Strategy Selection:
  • Heterosis: The significant dominance effects suggest that crossbreeding distinct lines could exploit heterosis (hybrid vigor) to rapidly improve productivity.
  • Diet-Specific vs. Combined: Due to moderate G × D interactions, breeders must decide between a single "generalist" breeding program (using combined data) or developing specific lines for specific waste streams (e.g., a line optimized for FVW).
• Selection Traits: Larval Surface Area (LSA) is a viable proxy for Body Weight (LBW), facilitating the use of automated computer vision systems for high-throughput phenotyping.

For the Circular Bioeconomy:

• The study provides the scientific foundation to industrialize BSF production. By optimizing genetics for specific waste streams, producers can maximize the bioconversion efficiency of diverse organic by-products, enhancing the economic viability of insect farming as a sustainable protein source.

Future Directions:

• The authors suggest that while individual genomic selection is currently impractical in mass-rearing, genomic prediction can be used to rank breeding lines. Future work should expand to include traits like protein/fat content and reproductive efficiency.