Bernard Steve Baleba Soh

Machine learning is transforming chemical ecology by accelerating the discovery and deployment of semiochemical-based tools for precision pest management. These advances are particularly important in the face of climate change, pesticide resistance, and the growing need for sustainable agricultural intensification. This review synthesizes how machine learning can be applied across the semiochemical discovery and implementation pipeline, from chemical signal detection to field deployment and decision support for integrated pest management. We review major machine learning approaches and demonstrate how they extract biologically relevant information from high-dimensional chemical, electrophysiological, behavioral, sensor, and field datasets. These methods accelerate semiochemical discovery, prioritize candidate compounds, optimize formulations and deployment strategies, and support adaptive pest management under dynamic environmental conditions. We further examine the integration of sensor-based technologies, explainable machine learning, and optimization algorithms to improve pest detection, monitoring, and intervention. Current challenges include data scarcity, heterogeneous datasets, the lack of benchmark resources, limited field validation, and the underrepresentation of tropical agroecosystems and smallholder farming systems. We propose practical recommendations for developing reusable datasets that integrate volatile profiles, electrophysiology, insect behavior, and management outcomes to improve model robustness and reproducibility. Central to this review is a 4-stage framework comprising detection, design, deployment, and decision support that connects machine learning with chemical ecology and provides a practical roadmap for translating computational advances into targeted, efficient, and sustainable pest management.

Terrestrial insects underpin key ecosystem services, including pollination, herbivory regulation, decomposition, nutrient cycling, and disease control. These functions depend on chemical communication that guides insects to food, mates, hosts, shelters, and oviposition sites while helping them avoid threats. Environmental microparticles, such as micro‐ and nanoplastics, tyre wear particles, soot, mineral dust, and agricultural residues, are now widespread across air, soil, vegetation, and indoor environments, exposing insects through contact, deposition, and ingestion. Growing evidence shows that these particles disrupt insect olfaction by adsorbing volatile compounds, blocking antennal sensilla, and interfering with receptor and neuronal processes. These disruptions impair foraging, mating, oviposition, and host seeking, leading to reduced individual performance and declines in population density. As a result, insect‐mediated services such as pollination, biological control, seed dispersal, and nutrient cycling are weakened, with consequences for plant community composition, biodiversity, food web stability, and crop yields. Despite these advances, the mechanisms linking microparticles to olfactory disruption remain poorly resolved, and no synthesis has integrated evidence across the full pathway from signal emission to neural processing. This review brings together chemical, physiological, and ecological evidence to map how microparticles interfere with insect olfaction across this pathway, identify key knowledge gaps, and propose a multi‐level research framework to guide future work from molecular mechanisms to ecological consequences.

Filth flies are groups of dipteran species closely associated with human and animal faeces, food waste, or carrion. They include species such as the house fly (Musca domestica), face fly (Musca autumnalis), bazaar fly (Musca sorbens), false stable flies (Muscina stabulans, Muscina levida), lesser house fly (Fannia canicularis), stable flies (Stomoxys spp.), horn fly (Haematobia irritans), neglected blood feeder fly (Musca crassirostris), oriental latrine fly (Chrysomya megacephala), green bottle fly (Lucilia sericata), Australian sheep blowfly (Lucilia cuprina) and blue bottle fly (Calliphora vicina). Thriving on decomposing organic waste, sewage, animal faeces, and carrion, these flies act as mechanical vectors of viruses, bacteria, protozoa, and helminths. They also cause direct harm through annoyance, stress, skin lesions, blood loss, and immunosuppression. In humans, contact with contaminated items, body parts, or food results in diseases such as trachoma, chlamydia, cholera, and gastrointestinal infections. In livestock, infestations lead to reduced weight gain and milk yield, causing economic losses exceeding USD 2 billion annually. Like all insects, filth flies rely heavily on olfaction to locate food, mates, oviposition sites, shelters, and to avoid danger. Identifying the odour cues that guide these behaviours offers promising opportunities for control, particularly when integrated into trapping or repellent systems. This review summarises current knowledge on filth fly management using odour cues. It explains the nature of these cues and their detection mechanisms in insects, discusses studies applying them for monitoring, attraction, repellency, or mass trapping, and highlights key considerations for developing effective odour-based control strategies. The review provides a useful reference for veterinarians, entomologists, chemical ecologists, industry, and funding agencies seeking sustainable alternatives for managing filth flies.

The success of invasive species often relies on sensory adaptations that allow them to exploit new environments. Unlike most of its relatives, the spotted-wing drosophila ( Drosophila suzukii ) lays eggs in ripening fruit, making it a serious and rapidly spreading agricultural pest worldwide. Spice plants produce volatiles that repel many insects, presenting potential sources of behavioural modulators for D. suzukii . Here, we examined whether volatiles from clove (Syzygium aromaticum) affect adult behaviour and offspring performance in D. suzukii . Behavioural assays revealed strong aversion: adults avoided grapes treated with clove extract, and gravid females laid significantly fewer eggs on treated fruit in both no-choice and choice assays. Clove exposure also reduced fitness, resulting in slower larval growth, lower adult weight, decreased emergence rate, and shortened lifespan. Gas chromatography–mass spectrometry analyses identified eugenol as the predominant volatile. Synthetic eugenol elicited strong avoidance across multiple host fruits, including apple, banana, blueberry, guava, mango, orange, papaya, raspberry, strawberry, and tomato. Molecular docking and single-sensillum recordings indicated that eugenol binds most strongly to DsuzOR88a and DsuzOR69aA odorant receptors. Cheminformatic screening of 1000 eugenol analogues revealed several structural variants predicted to interact strongly with OR88a, providing additional candidate repellents. By integrating behavioural, analytical chemistry, electrophysiology, and cheminformatics, our study shows that clove, through its major volatile eugenol, modulates host choice and reduces offspring performance in D. suzukii . These results highlight the ecological role of plant-derived volatiles in insect–plant interactions and support eugenol as a promising natural compound for sustainable fruit pest management.

This study investigated the effect of nutrition on development of wings and adult body condition of Bactrocera dorsalis (Hendel), an exotic pest of Asian origin and a species of economic importance in South Africa. Six host fruit types were used to rear B. dorsalis: mango, guava, orange, grapefruit, papaya and apple. One day after emergence, adults were used for tethered flight assessments using computerised flight mills. Additional flies were assessed in terms of wing morphology using geometric morphometrics, flight muscle mass, and lipid content. There were significant differences in the larval and pupal developmental time as well as larval and pupal survival amongst the tested host fruit, with orange tending to lead to poor performance in development and survival. Host fruit type and sex had a significant effect on the flight performance and wing morphology of B. dorsalis. There was a significant difference between female and male wing shapes, indicating the presence of sexual dimorphism. Females were more likely to fly than the males, and flies that developed on grapefruit covered the shortest distance. Host fruit type influenced both the lipid content and the flight muscle mass, with development in grapefruit leading to the lowest values. Our results show that variation in flight performance of B. dorsalis individuals can be associated with differences in resource quality during their development, and may lead to regional and temporal differences in the ability of the pest to disperse.

The success of invasive species often relies on sensory adaptations that allow them to exploit new environments. Unlike most of its relatives, the spotted wing drosophila (Drosophila suzukii) lays eggs in ripening fruit, making it a serious and rapidly spreading agricultural pest worldwide. Spice plants produce volatiles that repel many insects, making them a potential source of D. suzukii behavioural modulators. Here, we examined whether volatiles from clove (Syzygium aromaticum) could alter the behaviour and fitness of D. suzukii. Behavioural assays revealed strong aversion: adults avoided grapes treated with clove extract, while gravid females laid significantly fewer eggs on treated fruit in both no choice and choice assays. Clove exposure adversely impacted fitness, resulting in slower larval growth, reduced adult weight, decreased emergence rate, and shortened lifespan. Gas Chromatography mass spectrometry analyses identified eugenol as the predominant volatile. Consistent with this, synthetic eugenol elicited strong avoidance not only in grapes but also across various host fruits, including apple, banana, blueberry, guava, mango, orange, papaya, raspberry, strawberry, and tomato. Furthermore, molecular docking and single-sensillum recordings indicated that eugenol binds most strongly to DsuzOR88a, followed by DsuzOR69a and DsuzOR23a odorant receptors. Subsequently, cheminformatic screening of 1,000 eugenol analogues revealed several structural variants predicted to interact strongly with OR88a, offering additional candidate repellents. By combining behavioural, analytical chemistry, electrophysiological, and cheminformatic approaches, our study demonstrates that clove, through its major volatile eugenol, disrupts host use and reduces fitness in D. suzukii. These results highlight the ecological role of plant-derived repellents in insect plant interactions and position eugenol as a promising natural compound for sustainable fruit pest management.

In a world grappling with climate change, understanding the enduring impact of changes in temperature on insect adult traits is crucial. Here, we explored the intricate dynamics of exposure to different temperatures in three Drosophila species: Drosophila ezoana originating in Arctic regions, D. novamexicana in arid, hot environments, and in the cosmopolitan species D. virilis . Rearing these flies at 15, 20, 25, and 30°C revealed striking variations in their cuticular hydrocarbon (CHC) profiles, known to mediate mate recognition and prevent water loss in insects. The cold-adapted D. ezoana consistently exhibited reduced CHC levels with increasing temperatures, while the warm-adapted D. novamexicana and the cosmopolitan D. virilis displayed more nuanced responses. Additionally, we observed a significant influence of rearing temperature on the mating behavior of these flies, where those reared at the extreme tempreatures, 15 and 30°C, exhibiting reduced mating success. Consequently, this led to a decrease in the production of adult offspring. Also these adult offspring underwent notable alterations in life history traits, reaching adulthood more rapidly at 25 and 30°C but with lower weight and reduced longevity. Furthermore, among these offspring, those produced by the cold-adapted D. ezoana were more vulnerable to desiccation and starvation compared to those from the warm-adapted D. novamexicana and the cosmopolitan D. virilis . In summary, our research underscores the intricate interplay between temperature, ecological adaptation and various life history traits in Drosophila species from distinct agro-ecological regions. The observed variations in CHC profiles, mating behavior, fertility and responses to environmental stressors collectively provide valuable insights into how environmental conditions shape the biology and ecology of insects.

Rapid and ongoing climate change increases global temperature, impacts feeding, and reproduction in insects. The olfaction plays an important underlying role in these behaviors in most insect species. Here, we investigated how changing temperatures affect odor detection and ensuing behavior in three drosophilid flies: Drosophila novamexicana, D. virilis and D. ezoana, species adapted to life in desert, global, and subarctic climates, respectively. Using a series of thermal preference assays, we confirmed that the three species indeed exhibit distinct temperature preferences. Next, using single sensillum recording technique, we classified olfactory sensory neurons (OSNs) present in basiconic sensilla on the antenna of the three species and thereby identified ligands for each OSN type. In a series of trap assays we proceeded to establish the behavioral valence of the best ligands and chose guaiacol, methyl salicylate and isopropyl benzoate as representatives of a repellent, attractant and neutral odor. Next, we assessed the behavioral valence of these three odors in all three species across a thermal range (10-35 °C), with flies reared at 18 °C and 25 °C. We found that both developmental and experimental temperatures affected the behavioral performance of the flies. Our study thus reveals temperature-dependent changes in odor-guided behavior in drosophilid flies.