Asking the Population Directly: Molecular Resistance Diagnostics and Three Questions Manitoba Has Not Asked
Four separate articles in this journal have ended at the same sentence. The methods for answering those questions are mature, the samples pass through our hands weekly, and pooling makes the cost almost trivial
Abstract
This journal has now identified the same gap four times. Bed bug kdr status, rodent VKORC1 genotype, phosphine rph2 frequency in stored product beetles, and domestic cockroach allergen levels are all unmeasured in Manitoba, and in each case the relevant method is established elsewhere. This paper examines the diagnostic methodology itself and argues that the barrier is organisational rather than technical or financial. We set out the complementary strengths of bioassays and molecular assays, noting that bioassays detect the phenotype regardless of mechanism but require large numbers of live insects and are insensitive to low allele frequencies, while molecular assays detect resistance alleles at an early stage before phenotypic effects appear and identify the underlying mechanism. We then examine the development that makes population-scale screening affordable: pooled genotyping, in which allele frequencies are measured from pools of ten individuals with accuracy equivalent to genotyping them singly, at substantially lower time and cost, and with sensitivity sufficient to detect one heterozygote among nine wild-type individuals. We review recent applications across aphids, whiteflies and malaria vectors, including a case where molecular screening and bioassay gave systematically different pictures of the same populations, and we close with a concrete specification of what a Manitoba survey would involve.
1. Introduction: the same sentence four times
Writing this journal has produced an unexpected pattern. Four separate articles, on unrelated subjects, have arrived at the same closing observation: the relevant measurement exists as an established method, it has been applied elsewhere, and nobody has applied it here.
Bed bug kdr genotyping. Rodent VKORC1 status. Phosphine resistance in prairie stored product beetles. Domestic cockroach allergen concentrations. In each case we could describe what is known from populations studied elsewhere and could say nothing about Manitoba.
What this paper argues The obstacle is not that the science is hard, the instruments are exotic, or the samples are difficult to obtain. Pooled molecular screening has made this kind of survey cheap. The obstacle is that no one has convened it, which is a different kind of problem and a more solvable one.
2. What a bioassay measures
The traditional approach exposes live insects to a discriminating dose and counts survivors.
Bioassays may reliably detect resistance phenotypes at the population level regardless of the underlying cause of resistance.1 That property is genuinely valuable and molecular methods cannot match it, because a bioassay does not need to know why an insect survived.
2.1 The costs
Bioassays require high numbers of live insects, and are not very sensitive to low frequencies of resistance alleles or recessive mutations.1
The live requirement is a substantial practical constraint. Feasibility in high throughput depends on whether insects can be easily collected, stored and grown in the laboratory.3
For structural pest work this is limiting in a specific way. A technician collects dead insects routinely. Collecting, transporting and maintaining live colonies from field sites is a different undertaking that almost no commercial operation is equipped for.
2.2 The sensitivity problem
The insensitivity to low allele frequency matters more than it sounds. A population at five per cent resistance allele frequency will produce a bioassay result close to fully susceptible, because nearly every tested insect dies. The programme is told everything is fine during precisely the period when intervention would be cheapest.
3. What a molecular assay measures
The alternative asks the genome rather than the phenotype.
Molecular assays are applied either on pooled samples in the case of metabolic resistance, using gene expression techniques including reverse transcription quantitative PCR, microarrays and RNA sequencing, or on individual insects. For kdr genotyping the available options include allele-specific PCR, sequence-specific oligonucleotide probe assays and TaqMan assays.1
3.1 The two classes of target
The distinction matters. Target-site resistance, such as kdr or VKORC1, is a discrete point mutation and can be genotyped directly. Metabolic resistance is usually over-expression of a detoxifying enzyme, which is a quantitative change in gene expression rather than a change in sequence, and requires expression analysis instead.
This maps onto the resistance architecture described elsewhere in this journal. The four-layer bed bug defence includes both kinds, which means a complete picture requires both approaches.
3.2 The specific assay chemistry
TaqMan assays have been optimised to distinguish wild-type L1014 from the kdr point mutations 1014F and 1014S in a triplex reaction, and wild-type N1575 from the 1575Y mutation in a duplex reaction.1 The most widely used platform for estimating allele frequency from pooled DNA is the 5 prime nuclease assay, using TaqMan probes carrying a minor groove binding molecule and a fluorescent dye.5
This is standard molecular biology available in any reasonably equipped university laboratory.
4. Detecting resistance before it appears
The strongest argument for molecular monitoring is temporal.
Molecular assays complement bioassays in that they may detect resistant alleles at low frequencies at an early stage of resistance development, meaning incipient resistance, when the effect on the phenotype is not yet apparent. They also provide information on the underlying mechanism of resistance.1
Why early detection is the whole point By the time a bioassay records resistance, the allele is common enough to be failing in the field. A molecular assay can find it while it is rare, which is the only period during which rotation away from the affected mode of action would actually preserve it.
4.1 Connecting this to the dominance findings
This journal has repeatedly encountered resistance traits expressed in heterozygotes: glucose aversion is partially dominant, kdr in bed bugs affects knockdown even when heterozygous, and the strong phosphine phenotype is expressed in heterozygosity as well as homozygosity.
Dominant expression means selection acts immediately, so the window between first appearance and field failure is short. A monitoring method that only detects the problem at the end of that window is providing information too late to act on.
5. The pooling breakthrough
The development that changes the economics is unglamorous and decisive.
Individual genotyping by PCR-RFLP is labour intensive and limits the number of sites that can be monitored, offering limited benefit over the traditional bioassay. Cost-effective methods to monitor resistance allele frequencies in field populations are critical to maintaining successful integrated pest management.5
5.1 The result
A pooled approach is able to measure allele frequencies from pools of ten mosquitoes with very high accuracy, equivalent to genotyping individuals, while requiring substantially less time and lower cost.1
The validation figures are specific. The method was found to be precise at 1.66 to 2.99 per cent and accurate at 3.3 to 5.9 per cent, and was able to detect a single heterozygous individual mixed with nine wild-type individuals in a pool of ten. Pilot application to field-caught samples showed minimal differences from individual genotyping, at 0.36 to 4.0 per cent.1
5.2 Why this matters for a provincial survey
Tenfold reduction in assay count at negligible loss of accuracy converts a research project into an affordable exercise. A survey of 300 insects becomes 30 reactions.
The sensitivity figure is the one that removes the obvious objection. A method that can find one heterozygote in ten is capable of detecting incipient resistance rather than only established resistance, which is precisely the capability §4 identified as the point of the exercise.
The approach demonstrated real-world value in exactly this way: it allowed the first detection of the super-kdr mutation N1575Y in Anopheles gambiae from Mali.1 Using pools instead of individuals allows more efficient resistance allele screening.1
6. When the two methods disagree
A study of Australian whitefly populations provides the clearest illustration of why both methods are needed, because the two gave systematically different pictures.
Researchers used metabarcoding to analyse species composition across 144 field populations collected between 2013 and 2021, then applied high-throughput sequencing of organophosphate and pyrethroid resistance gene amplicons.4
6.1 The molecular picture
The organophosphate resistance allele F331W was fixed, above 99 per cent, in all MEAM1 populations. The pyrethroid resistance allele L925I in the voltage-gated sodium channel gene was detected at varying frequencies: 1.0 to 7.0 per cent in 43 populations, 27.7 and 42.1 per cent in two populations, and 95 to 97.5 per cent in three populations. Neither allele was detected in the AUS I populations.4
6.2 The bioassay picture
Pyrethroid bioassays of 85 field-derived populations detected no resistance in 51 populations, low frequency resistance in 32, and high resistance in two.4
6.3 Reading them together
Fifty one populations looked clean by bioassay. The sequencing found the pyrethroid allele present at low frequency across 43 populations, and an organophosphate allele fixed everywhere.
Neither method was wrong. The bioassay correctly reported that pyrethroids still worked in those populations. The sequencing correctly reported that the allele was already circulating. A programme relying on bioassay alone would have had no warning; a programme relying on genotype alone might have abandoned a chemistry that was still effective.
7. Panels rather than single markers
Contemporary practice has moved from testing one mutation to testing many at once.
A study of Greek aphid populations collected between 2021 and 2025 combined diagnostic bioassays across five insecticides with a newly developed, highly sensitive droplet digital PCR panel targeting seven key resistance mutations, spanning pyrethroid resistance through voltage-gated sodium channel mutations, keto-enol resistance, dimethyl carbamate resistance, and resistance to nicotinic acetylcholine receptor competitive modulators linked to cytochrome P450 over-expression.2
The ddPCR analysis, conducted on 634 aphids, confirmed the presence of six of the seven mutations.2
7.1 The bioassay findings alongside
The parallel bioassays revealed frequent cross-resistance, reported the first case of resistance to one active, and produced a resistance frequency spread across products ranging from 50.0 per cent for one to 1.7 per cent for another.2
7.2 Why panels are the right design
A panel answers the operational question directly. A practitioner does not want to know whether a population is resistant in the abstract. They want to know which of the available products will still work, and a multi-marker panel returns something close to that.
The agricultural literature anticipates this direction, noting that as an increasing number of molecular markers are identified, high-throughput, fast and accurate molecular diagnostic platforms could overcome the need for time-consuming bioassays.3 Section 8 argues for a more cautious version of that conclusion.
8. The complementarity conclusion
The position we think the evidence supports is stated plainly in the whitefly study.
The authors conclude that high-throughput sequencing and bioassays are complementary approaches, and that detection of target site mutations together with phenotypic data provides a comprehensive analysis.4
The malaria vector literature reaches the same place, describing a suite of bioassays including diagnostic dose and intensity assays and synergist use, alongside complementary assays for known molecular markers.1
8.1 Why we resist the replacement framing
The suggestion that molecular platforms could overcome the need for bioassays3 is attractive and, we think, premature for structural pest management specifically.
The reason is §9. Molecular assays find the mutations someone has already characterised, and in several of the systems this journal has examined, resistance is layered across mechanisms that are not all reducible to genotyped point mutations.
9. The limits of molecular methods
An honest account has to include what genotyping cannot do.
It only finds what is already known. An assay targets a specific characterised mutation. A population resistant by an undescribed mechanism genotypes as susceptible.
The discovery pipeline is separate and expensive. Identifying new markers begins with observing decreased sensitivity in field populations and proceeds through whole genome genotyping of resistant and susceptible individuals, comparing allele frequencies between groups.6 Large-scale genotyping-based candidate studies are limited, likely because of the difficulty of identifying convincing candidates and the high cost of focal genotyping at appropriate scale.6
Metabolic resistance is harder. It requires expression analysis rather than genotyping,1 which is more demanding on sample handling since RNA degrades quickly.
Behavioural resistance is invisible to it entirely. As set out in the glucose aversion article published here, that trait acts before ingestion and would not appear in any genotyping panel targeting toxicant metabolism or target sites.
Genotype is not phenotype. Allele frequency predicts field performance only through a relationship that has to be established empirically, which is the bioassay's job.
10. Three Manitoba questions
Bringing this back to the gap identified in §1.
Each of these has the same structure. A named mutation, characterised in the published literature. An established assay. A species present in Manitoba. Samples that pass through the hands of pest management operators routinely. And no published local data.
10.1 Why these three specifically
Bed bug kdr determines whether pyrethroid products retain any value in local multi-unit housing, which as set out elsewhere in this journal is where the problem concentrates.
Rodent VKORC1 determines which anticoagulants remain useful. Because the characterised mutations confer resistance to all first-generation compounds and to some second generation compounds but not others, the specific allele present determines the specific product choice.
Stored product rph2 matters because phosphine is the only general use fumigant and because roughly three quarters of surveyed populations globally are resistant, with the relevant species established on the prairie.
The fourth gap, domestic allergen levels, requires dust sampling and immunoassay rather than genotyping and sits outside this paper's scope, though it is equally tractable.
11. What a survey would actually involve
A concrete specification, since vague calls for research are easy to make and easy to ignore.
Samples. Dead specimens from routine service work, which requires no change to anyone's practice beyond retaining and labelling them. Rodent tail tips, as used in the published VKORC1 surveys. Bed bugs from confirmed infestations. Beetles from grain facilities.
Metadata. Location at postal-code resolution, building type, and treatment history where known. Without this the allele frequencies are uninterpretable.
Scale. Pooling at ten reduces a few hundred specimens per species to a few dozen reactions.1
Assay. Published primers exist for all three markers. TaqMan or ddPCR platforms are standard university equipment.12
Bioassay subset. Per §8, a smaller live-insect component on selected populations to anchor genotype to phenotype.
Publication. Results public, since the value is in operators and regulators being able to act on them.
11.1 What it would answer
Which products still work here, which are being wasted, and whether the province is early or late in a resistance trajectory that is well advanced elsewhere.
12. Why it has not happened
Our assessment, offered as analysis rather than finding.
The benefit is collective and the cost is individual. No single operator captures enough of the value to fund it, which is the structure of a public good and the reason such things are usually publicly funded.
The parties do not overlap. Operators have the samples and no laboratory. University laboratories have the capability and no samples. Neither has a mechanism for finding the other.
The result may be commercially uncomfortable. A survey establishing that widely sold products do not work locally is not obviously in the interest of everyone selling them.
It is nobody's job. Pesticide regulation is federal, pest control licensing is provincial, and public health surveillance covers disease rather than resistance. Resistance monitoring falls between all three.
None of these is insurmountable. They are, collectively, why something cheap and useful remains undone.
13. Limitations and open questions
The methods literature is agricultural and vector-focused. Our sources concern mosquitoes, aphids and whiteflies.124 The molecular techniques transfer directly, but sampling design for structural pests in buildings differs from field sampling and would need work.
Pooling validation is species-specific. The pool-of-ten accuracy figures were established in Anopheles.1 Applying the approach to bed bugs or rodents would require its own validation.
We have not costed it. We assert affordability on the strength of the tenfold reduction in reaction count.1 We have not obtained quotations, and a serious proposal would need to.
Section 12 is our reasoning. The explanations for why the work has not been done are ours and are not sourced.
We would benefit. Knowing local resistance status would improve our own product selection. It would also potentially establish that things we currently sell do not work, and we would rather know.
14. Conclusion
Bioassays detect the resistance phenotype whatever its cause but need large numbers of live insects and miss low allele frequencies.1 Molecular assays detect resistance alleles early, before phenotypic effects appear, and identify the mechanism.1 The published position, which we adopt, is that they are complementary rather than substitutes.4
Pooled genotyping has made population screening cheap. Allele frequencies can be measured from pools of ten with accuracy equivalent to individual genotyping, at substantially lower time and cost, with sensitivity sufficient to find one heterozygote among nine wild-type individuals.1 Contemporary panels test seven mutations at once across hundreds of specimens.2
Manitoba has three characterised markers of immediate practical relevance, a species present for each, published assays, and specimens passing through operators' hands every week. What it does not have is anyone whose job it is to connect those facts.
Four articles in this journal have now ended by noting that a question has not been asked here. This one is about the fact that asking would be straightforward, and that the reason it remains unasked is organisational rather than scientific. That is the kind of problem a province can fix in an afternoon of deciding to.
References
- Detection and Monitoring of Insecticide Resistance Mutations in Anopheles gambiae: Individual vs. Pooled Specimens. Genes, 9(10), 479 (2018). doi:10.3390/genes9100479. Source for the comparison of bioassay and molecular diagnostic strengths, the requirement for high numbers of live insects and insensitivity to low allele frequencies, the detection of incipient resistance and mechanism identification by molecular assays, the TaqMan triplex and duplex assay designs for L1014F, L1014S and N1575Y, the pooled approach measuring allele frequencies from pools of ten with precision of 1.66 to 2.99 per cent and accuracy of 3.3 to 5.9 per cent, detection of a single heterozygote among nine wild-type individuals, field agreement of 0.36 to 4.0 per cent with individual genotyping, and the first detection of N1575Y in Mali. https://pmc.ncbi.nlm.nih.gov/articles/PMC6209882/
- Mavridis and colleagues. Molecular droplet digital PCR diagnostics and bioassays for monitoring insecticide resistance status in Myzus persicae populations from Greece. Pest Management Science. doi:10.1002/ps.70743. Source for the ddPCR panel targeting seven resistance mutations across voltage-gated sodium channel, acetyl-CoA carboxylase, acetylcholinesterase and nicotinic acetylcholine receptor targets, the analysis of 634 aphids confirming six mutations, the parallel bioassays across five insecticides revealing frequent cross-resistance and a first case of resistance to one active, and the resistance frequency range from 50.0 per cent to 1.7 per cent across products. https://scijournals.onlinelibrary.wiley.com/doi/10.1002/ps.70743
- Significance and interpretation of molecular diagnostics for insecticide resistance management of agricultural pests. Current Opinion in Insect Science. Source for the dependence of bioassay throughput on ease of collection, storage and laboratory rearing, and for the argument that high-throughput molecular platforms could overcome the need for time-consuming bioassays as marker numbers increase. https://www.sciencedirect.com/science/article/abs/pii/S2214574520300432
- Screening for insecticide resistance in Australian field populations of Bemisia tabaci using bioassays and DNA sequencing. Pest Management Science. Source for the metabarcoding of 144 field populations collected 2013 to 2021, the F331W organophosphate allele fixed above 99 per cent in all MEAM1 populations, the L925I pyrethroid allele frequency distribution across populations, the bioassay results across 85 populations with no resistance detected in 51, and the conclusion that high-throughput sequencing and bioassays are complementary approaches. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9546000/
- Quantification of the Pirimicarb Resistance Allele Frequency in Pooled Cotton Aphid (Aphis gossypii Glover) Samples by TaqMan SNP Genotyping Assay. Source for the limitations of individual genotyping by PCR-RFLP as labour intensive with limited benefit over traditional bioassay, the need for cost-effective resistance allele frequency monitoring, and the description of the 5 prime nuclease assay with minor groove binding TaqMan probes as the most widely used platform for pooled allele frequency estimation. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3948748/
- Identification, validation and application of molecular diagnostics for insecticide resistance in malaria vectors. Source for the marker discovery pipeline beginning from observed phenotypic insensitivity and proceeding through whole genome genotyping and allele frequency comparison, and for the observation that large-scale genotyping-based candidate studies are limited by difficulty identifying convincing candidates and by the cost of focal genotyping at scale. https://pmc.ncbi.nlm.nih.gov/articles/PMC4767538/
How to cite this article
APC Exterminators Research Division (2026). Asking the Population Directly: Molecular Resistance Diagnostics and Three Questions Manitoba Has Not Asked. APC Review, Data, Statistics & Bioinformatics. Retrieved from https://apcexterminators.com/insights/molecular-resistance-diagnostics-pooled-genotyping-manitoba-gap