Mosquito Identification Research Articles

Enhancing mosquito classification through self-supervised learning

Traditional mosquito identification methods, relied on microscopic observation and morphological characteristics, often require significant expertise and experience, which can limit their effectiveness. This study introduces a self-supervised learning-based image classification model using the Bootstrap Your Own Latent (BYOL) algorithm, designed to enhance mosquito species identification efficiently. The BYOL algorithm offers a key advantage by eliminating the need for labeled data during pretraining, as it autonomously learns important features. During fine-tuning, the model requires only a small fraction of labeled data to achieve accurate results. Our approach demonstrates impressive performance, achieving over 96.77% accuracy in mosquito image analysis, with minimized both false positives and false negatives. Additionally, the model’s overall accuracy, measured by the area under the ROC curve, surpasses 99.55%, highlighting its robustness and reliability. A notable finding is that fine-tuning with just 10% of labeled data produces results comparable to using the full dataset. This is particularly valuable for resource-limited settings with limited access to advanced equipment and expertise. Our model provides a practical solution for mosquito identification, overcoming the challenges of traditional microscopic methods, such as the time-consuming process and reliance on specialized knowledge in healthcare services. Overall, this model supports personnel in resource-constrained environments by facilitating mosquito vector density analysis and paving the way for future mosquito species identification methodologies.

DNA Barcoding for the Identification of Adult Mosquitoes (Diptera: Culicidae) in Western Australia.

Precise mosquito identification is integral to effective arbovirus surveillance. Nonetheless, the conventional morphological approach to identifying mosquito species is laborious, demands expertise and presents challenges when specimens are damaged. DNA barcoding offers a promising alternative, surmounting challenges inherent in morphological identification. To integrate DNA barcoding into arbovirus surveillance effectively, a robust dataset of mosquito barcode sequences is required. This study established a comprehensive repository of Cytochrome Oxidase I (COI) barcodes, encompassing 177 samples representing 45 mosquito species from southern and northern Western Australia (WA), including 16 species which have not been previously barcoded. The average intraspecific and interspecific genetic distances were 1% and 6.8%, respectively. Anopheles annulipes sensu lato had the highest intraspecific distance at 9.1%, signifying a genetically diverse species. While validating the potential of COI barcodes to accurately differentiate mosquito species, we identified that some species pairs have low COI divergence. This includes Aedes clelandi and Ae. hesperonotius, Tripteroides atripes and Tp. punctolaeralis and Ae. turneri and Ae. stricklandi. In addition, we observed ambiguity in identification of the members of Culex sitiens subgroup (Cx. annulirostris, Cx. palpalis and Cx. sitiens) and three members of Cx. pipiens complex (Cx. australicus, Cx. globocoxitus, Cx. quinquefasciatus). In summary, despite presenting challenges in the identification of some mosquito species, the COI barcode accurately identified most of the species and generated a valuable resource that will support the WA arbovirus surveillance program and enhance public health intervention strategies for mosquito-borne disease control.

Towards transforming malaria vector surveillance using VectorBrain: a novel convolutional neural network for mosquito species, sex, and abdomen status identifications

Malaria is a major public health concern, causing significant morbidity and mortality globally. Monitoring the local population density and diversity of the vectors transmitting malaria is critical to implementing targeted control strategies. However, the current manual identification of mosquitoes is a time-consuming and intensive task, posing challenges in low-resource areas like sub-Saharan Africa; in addition, existing automated identification methods lack scalability, mobile deployability, and field-test validity. To address these bottlenecks, a mosquito image database with fresh wild-caught specimens using basic smartphones is introduced, and we present a novel CNN-based architecture, VectorBrain, designed for identifying the species, sex, and abdomen status of a mosquito concurrently while being efficient and lightweight in computation and size. Overall, our proposed approach achieves 94.44±2% accuracy with a macro-averaged F1 score of 94.10±2% for the species classification, 97.66±1% accuracy with a macro-averaged F1 score of 96.17±1% for the sex classification, and 82.20±3.1% accuracy with a macro-averaged F1 score of 81.17±3% for the abdominal status classification. VectorBrain running on local mobile devices, paired with a low-cost handheld imaging tool, is promising in transforming the mosquito vector surveillance programs by reducing the burden of expertise required and facilitating timely response based on accurate monitoring.

Barcoding of Italian mosquitoes (BITMO): generation and validation of DNA barcoding reference libraries for native and alien species of Culicidae

BackgroundMosquitoes (Culicidae), as disease vectors, represent a risk for human health worldwide. Repeated introductions of alien mosquito species and the spread of invasive species have been recorded in different countries. Traditionally, identification of mosquitoes relies on morphological observation. However, morphology-based identification is associated with a number of potential disadvantages, such as the high level of specialisation of the operator and its limited applicability to damaged samples. In these cases, species identification is achieved through molecular methods based on DNA amplification. Molecular-based taxonomy has also enabled the development of techniques for the study of environmental DNA (eDNA). Previous studies indicated the 16S mitochondrial ribosomal RNA (rRNA) gene as a promising target for this application; however, 16S rRNA sequences are available for only a limited number of mosquito species. In addition, although primers for the 16S rRNA gene were designed years ago, they are based on limited numbers of mosquito sequences. Thus, the aims of this study were to: (i) design pan-mosquito 16S rRNA gene primers; (ii) using these primers, generate a 16S rRNA gene mosquito reference library (with a focus on mosquitoes present in Italy); and (iii) compare the discriminatory power of the 16S rRNA gene with two widely used molecular markers, cytochrome c oxidase subunit 1 mitochondrial gene (COI) and internal transcribed spacer 2 (ITS2).MethodsA total of six mosquito genera (28 mosquito species) were included in this study: Aedes (n = 16 species), Anopheles (5 species), Coquillettidia (1 species), Culex (3 species), Culiseta (2 species) and Uranotaenia (1 species). DNA was extracted from the whole mosquito body, and more than one specimen for each species was included in the analysis. Sanger sequencing was used to generate DNA sequences that were then analysed through the Barcode of Life Data Systems (BOLD). Phylogenetic analyses were also performed.ResultsNovel 16S rDNA gene, COI and ITS2 sequences were generated. The 16S rRNA gene was shown to possess sufficient informativeness for the identification of mosquito species, with a discriminatory power equivalent to that of COI.ConclusionsThis study contributes to the generation of DNA barcode libraries, focussed on Italian mosquitoes, with a significant increase in the number of 16S rRNA gene sequences. We hope that these novel sequences will provide a resource for studies on the biodiversity, monitoring and metabarcoding of mosquitoes, including eDNA-based approaches.Graphical abstract

Robust mosquito species identification from diverse body and wing images using deep learning

Mosquito-borne diseases are a major global health threat. Traditional morphological or molecular methods for identifying mosquito species often require specialized expertise or expensive laboratory equipment. The use of convolutional neural networks (CNNs) to identify mosquito species based on images may offer a promising alternative, but their practical implementation often remains limited. This study explores the applicability of CNNs in classifying mosquito species. It compares the efficacy of body and wing depictions across three image collection methods: a smartphone, macro-lens attached to a smartphone and a professional stereomicroscope. The study included 796 specimens of four morphologically similar Aedes species, Aedes aegypti, Ae. albopictus, Ae. koreicus and Ae. japonicus japonicus. The findings of this study indicate that CNN models demonstrate superior performance in wing-based classification 87.6% (95% CI: 84.2–91.0) compared to body-based classification 78.9% (95% CI: 77.7–80.0). Nevertheless, there are notable limitations of CNNs as they perform reliably across multiple devices only when trained specifically on those devices, resulting in an average decline of mean accuracy by 14%, even with extensive image augmentation. Additionally, we also estimate the required training data volume for effective classification, noting a reduced requirement for wing-based classification compared to body-based methods. Our study underscores the viability of both body and wing classification methods for mosquito species identification while emphasizing the need to address practical constraints in developing accessible classification systems.Graphical abstract

Biting Rhythms of Malaria Vector in the Urban Area of Itu, Akwa Ibom State, Nigeria

In Nigeria, malaria is still a serious public health concern. Studies on the endemicity of the illness in Akwa Ibom State have been conducted to identify effective strategies for malaria control, but little is known about the spread of Anopheles in the metropolitan Itu area. In the study area, a descriptive study was conducted between November 2017 and December 2018. The CDC light trap human baited bednet method was used to measure the biting activity of both indoor and outdoor mosquitoes, while the pyrethroid spray catch technique was utilized to estimate the density of mosquitoes resting indoors. Standard entomological instruments were used for morphological identification of mosquitoes. Out of the 1263 mosquitoes that were captured, An. gambiae s l (87.49%) was more prevalent. The wet season had a considerably (p<0.05) greater indoor resting density of Anopheles mosquitoes than the dry season. The average indoor resting density was 39.40 Anopheles/room, and the annual man-biting rate was a considerably (p <0.05) high 59.38 bites/man. The peak biting times were seen to be between 2 and 3 a.m., indoors (23.97%) and between 12 and 1 a.m., outdoors (19.74%). Year-round transmission of malaria occurs. The primary malaria vector in Itu metropolitan region is Anopheles gambiae. Planning and carrying out malaria control measures is so crucial in the Itu Local Government Area.

Livestock keeping, mosquitoes and community viewpoints: a mixed methods assessment of relationships between livestock management, malaria vector biting risk and community perspectives in rural Tanzania

BackgroundLivestock keeping is one of the potential factors related to malaria transmission. To date, the impact of livestock keeping on malaria transmission remains inconclusive, as some studies suggest a zooprophylactic effect while others indicate a zoopotentiation effect. This study assessed the impact of livestock management on malaria transmission risks in rural Tanzania. Additionally, the study explored the knowledge and perceptions of residents about the relationships between livestock keeping and malaria transmission risks in a selected village.MethodsIn a longitudinal entomological study in Minepa village, South Eastern Tanzania, 40 households were randomly selected (20 with livestock, 20 without). Weekly mosquito collection was performed from January to April 2023. Indoor and outdoor collections used CDC-Light traps, Prokopack aspirators, human-baited double-net traps, and resting buckets. A subsample of mosquitoes was analysed using PCR and ELISA for mosquito species identification and blood meal detection. Livestock's impact on mosquito density was assessed using negative binomial GLMMs. Additionally, in-depth interviews explored community knowledge and perceptions of the relationship between livestock keeping and malaria transmission risks.ResultsA total of 48,677 female Anopheles mosquitoes were collected. Out of these, 89% were Anopheles gambiae sensu lato (s.l.) while other species were Anopheles funestus s.l., Anopheles pharoensis, Anopheles coustani, and Anopheles squamosus. The findings revealed a statistically significant increase in the overall number of An. gambiae s.l. outdoors (RR = 1.181, 95%CI 1.050–1.862, p = 0.043). Also, there was an increase of the mean number of An. funestus s.l. mosquitoes collected in households with livestock indoors (RR = 2.866, 95%CI: 1.471–5.582, p = 0.002) and outdoors (RR = 1.579,95%CI 1.080–2.865, p = 0.023). The human blood index of Anopheles arabiensis mosquitoes from houses with livestock was less than those without livestock (OR = 0.149, 95%CI 0.110–0.178, p < 0.001). The majority of participants in the in-depth interviews reported a perceived high density of mosquitoes in houses with livestock compared to houses without livestock.ConclusionDespite the potential for zooprophylaxis, this study indicates a higher malaria transmission risk in livestock-keeping communities. It is crucial to prioritize and implement targeted interventions to control vector populations within these communities. Furthermore, it is important to enhance community education and awareness regarding covariates such as livestock that influence malaria transmission.

Involving citizen scientists in monitoring arthropod vectors of human and zoonotic diseases: The case of Mosquito Alert in Italy

Citizen science has been particularly effective in gathering reliable, timely, large-scale data on the presence and distributions of animal species, including mosquito vectors of human and zoonotic pathogens. This involves the participation of citizen scientists in research projects, with success strongly dependent on the capacity to disseminate project information and engage citizen scientists to contribute their time. Mosquito Alert is a citizen science that aids in the system surveillances of vector mosquitoes. It involves citizen scientists providing expert-validated photos of targeted mosquitoes, along with records of bites and breeding sites. Since 2020 the system has been disseminated throughout Europe. This article uses models to analyze the effect of promotion activities carried out by the Mosquito Alert ITALIA team from October 2020 to December 2022 on the number of citizen scientists recruited and engaged in the project, and their performance in mosquito identification. Results show a high level of citizen scientist recruitment (N > 18.000; 37 % of overall European participants). This was achieved mostly through articles generated by ad hoc press releases detailing the app's goals and functioning. Press releases were more effective when carried out at the beginning and end of the mosquito season and when mosquito's public health significance was emphasized. Despite the high number of records received (N > 20.000), only 30 % of registered participants sent records, and the probability of a participant sending a record dropped off quickly over time after first registering. Among participants who contributed, ∼50 % sent 1 record, ∼30 % ≥3 and 4 % >10 records. Participants showed good capacity to identify mosquitoes and improve identification skills with app usage. The results will be valuable for anyone interested in evaluating citizen science, as participation and engagement are seldom quantitatively assessed. Our results are also useful for designing dissemination and education strategies in citizen science projects associated with arthropod vector monitoring.

Molecular characterization of Armigeres Subalbatus from Hyderabad region of Telangana state, India.

The mosquito Armigeres subalbatus (Coquillett, 1898) is a significant vector for Japanese encephalitis infection, and breeds in high organic polluted water. Understanding mosquito diversity and there abundance in relation to mosquito-borne diseases is an important component for public health managers. Though the conventional methods for systematic position of mosquito species by using morphological characteristics is a classical method, but it requires perfect expertise and well preserved specimen. Conversely, the molecular analysis of mitochondrial cytochrome C oxidase subunit I (COI) serves as a gene-centric DNA barcoding approach and offers a promising alternative method for mosquito species identification. The study at hand delves into the morphological characteristics of Armigeres subalbatus were compared with COI- gene to ensure a more dependable verification for identification of mosquito species found in Hyderabad region of Telangana. The 489 base pair amplicons were acquired and deposited into the NCBI Gene Bank nucleotide database under the accession number MG686500. The maximum likelihood tree infers that, the Hyderabad species was diverged from USA and Japan species but had ancestral relationship with Tamil Nadu, Karnataka, Maharastra, Kerala and Goa species. Mitochondrial gene (COI) based DNA barcoding is the most reliable and potential alternative technique to identify the mosquito species.

Identification of Southeast Asian Anopheles mosquito species using MALDI-TOF mass spectrometry.

Malaria elimination in Southeast Asia remains a challenge, underscoring the importance of accurately identifying malaria mosquitoes to understand transmission dynamics and improve vector control. Traditional methods such as morphological identification require extensive training and cannot distinguish between sibling species, while molecular approaches are costly for extensive screening. Matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS) has emerged as a rapid and cost-effective tool for Anopheles species identification, yet its current use is limited to few specialized laboratories. This study aimed to develop and validate an online reference database for MALDI-TOF MS identification of Southeast Asian Anopheles species. The database, constructed using the in-house data analysis pipeline MSI2 (Sorbonne University), comprised 2046 head mass spectra from 209 specimens collected at the Thailand-Myanmar border. Molecular identification via COI and ITS2 DNA barcodes enabled the identification of 20 sensu stricto species and 5 sibling species complexes. The high quality of the mass spectra was demonstrated by a MSI2 median score (min-max) of 61.62 (15.94-77.55) for correct answers, using the best result of four technical replicates of a test panel. Applying an identification threshold of 45, 93.9% (201/214) of the specimens were identified, with 98.5% (198/201) consistency with the molecular taxonomic assignment. In conclusion, MALDI-TOF MS holds promise for malaria mosquito identification and can be scaled up for entomological surveillance in Southeast Asia. The free online sharing of our database on the MSI2 platform (https://msi.happy-dev.fr/) represents an important step towards the broader use of MALDI-TOF MS in malaria vector surveillance.

Hybrid Machine Learning Approach for Mosquito Species Classification Using Wingbeat Analysis

Effective and precise techniques for mosquito species identification are required as mosquito-borne illnesses continue to pose serious threats to public health across the world. We provide a new hybrid machine-learning technique in this research work for the classification of mosquito species through the Wingbeat analysis. It analyzes the wingbeat of the mosquito species based on which it can identify the mosquito species. This method makes use of deep learning techniques. The hybrid technique attempts to provide robust and dependable classification performance by utilizing a wide range of machine learning methods, such as k-Nearest Neighbors (KNN), Random Forest, Multi-layer Perceptron (MLP), Support Vector Machines (SVM), and Gradient Boosting. To improve feature extraction and normalization, we apply a rigorous set of preprocessing techniques to a large dataset that includes wingbeat recordings from many mosquito species. By means of comprehensive testing and analysis, we prove that our method is effective in correctly detecting mosquito species, exhibiting better results than using separate machine learning algorithms. Our findings demonstrate how deep learning methods may support more conventional machine learning strategies in problems involving the categorization of mosquito species. We also address the implications of our results for ecological research and disease management initiatives, highlighting the significance of precise species identification in vector monitoring and epidemiological investigations.

Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry: An Innovative Tool for Rapid Identification of Hylurgus ligniperda, an Invasive Pest

Hylurgus ligniperda is an imported quarantine plant pest in China. Its identification is usually based on morphological characteristics; therefore, species identification needs high professional requirements of staff and professionals with high experience accumulated through long-term training. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is a rapid identification technology, which is based on protein profiles of species. It has been widely used for the identification of pathogenic microorganisms. Many studies have reported the identification of mosquitoes, ticks, and other arthropods. The application of MALDI-TOF MS in the identification of H. ligniperda can improve the identification efficiency of H. ligniperda, preventing and control its harm and further spread. To construct a spectra database for H. ligniperda, we analyzed the effect of different factors, such as different body parts, developmental stages, populations, and preservation conditions, on its protein spectrum. We collected protein spectrum profiles from 19 specimens of H. ligniperda and its related species, obtaining 211 protein spectra to construct a reference database and validate identification. The protein spectrum from the chest specimens of H. ligniperda showed many peaks, high intensity, and a stable signal, indicating a successful data establishment. The difference in protein spectra between different regions of the same species was less, but did not affect the identification results. Clear differences were observed in the protein spectrum across many developmental stages. The database established by the adult specimens protein spectrum can accurately identify Dendroctonus valens, Tomicus piniperda, and H. ligniperda. MALDI-TOF MS technology can be used for the rapid identification of H. ligniperda. This method is rapid and direct, and the identification results are robust. It does not require specialized entomological expertise and can be used for customs interception and field investigations.

NOVEL TRAP CONFIGURATION FOR LIVE CAPTURE OF MOSQUITOES

Diverse mosquito traps are available for mosquito and arbovirus surveillance. The delicate nature of the mosquito body makes them vulnerable to damage as they pass through the trap's fan, which can lead to rapid desiccation or mortality within the capture chamber. This can negatively affect surveillance accuracy, impacting both the precise identification of mosquitoes and the reliable execution of molecular assays for arbovirus detection. In this study, we report a novel modification to three widely used mosquito traps: CDC light trap, BG-Sentinel trap, and CDC gravid trap, incorporating a mesh funnel and updraft design to address these issues. We compared updraft and downdraft configurations of light traps under field conditions and compared the effectiveness of the modified BG and gravid trap to unaltered counterparts in semi-field environments. Subsequently, we conducted field validation of modified mosquito traps to assess their trapping effectiveness in terms of mosquito abundance and species composition in coastal forest and suburban areas. Our findings revealed that there was no significant difference in trapping effectiveness between different fan configurations. The adaptation made to the BG trap exhibited higher recapture rates of Culex quinquefasciatus and Aedes albopictus in comparison to the unmodified BG-Sentinel trap. The modification of the gravid trap was equivalent to unaltered CDC gravid trap. During our field trial, the modified light traps had a higher collection rate of mosquitoes with a wider spectrum of species diversity than modified BG or gravid traps, regardless of site. The modified BG traps captured more arbovirus vector species (Culex and Aedes species), with an increase in Ae. albopictus (11 times) and Ae. aegypti (1.75 times) when compared to the light traps. The modified gravid traps mostly collected Culex spp., accounting for over 47% of the collected mosquitoes. The results indicate that the novel trap configuration preserves trap functionality and improves specimen quality by avoiding the death and dismemberment of collected mosquitoes.

Accelerating targeted mosquito control efforts through mobile West Nile virus detection.

Different mosquito control strategies have been implemented to mitigate or prevent mosquito-related public health situations. Modern mosquito control largely relies on multiple approaches, including targeted, specific treatments. Given this, it is becoming increasingly important to supplement these activities with rapid and mobile diagnostic capacities for mosquito-borne diseases. We aimed to create and test the applicability of a rapid diagnostic system for West Nile virus that can be used under field conditions. In this pilot study, various types of adult mosquito traps were applied within the regular mosquito monitoring activity framework for mosquito control. Then, the captured specimens were used for the detection of West Nile virus RNA under field conditions with a portable qRT-PCR approach within 3-4h. Then, positive samples were subjected to confirmatory RT-PCR or NGS sequencing in the laboratory to obtain genome information of the virus. We implemented phylogenetic analysis to characterize circulating strains. A total of 356 mosquito individuals representing 7 species were processed in 54 pools, each containing up to 20 individuals. These pools were tested for the presence of West Nile virus, and two pools tested positive, containing specimens from the Culex pipiens and Anopheles atroparvus mosquito species. As a result of subsequent sequencing, we present the complete genome of West Nile virus and Bagaza virus. The rapid identification of infected mosquitoes is the most important component of quick response adulticide or larvicide treatments to prevent human cases. The conceptual framework of real-time surveillance can be optimized for other pathogens and situations not only in relation to West Nile virus. We present an early warning system for mosquito-borne diseases and demonstrate its application to aid rapid-response mosquito control actions.

Rapid detection of West Nile and Dengue viruses from mosquito saliva by loop-mediated isothermal amplification and displaced probes.

Arthropod-borne viruses are major causes of human and animal disease, especially in endemic low- and middle-income countries. Mosquito-borne pathogen surveillance is essential for risk assessment and vector control responses. Sentinel chicken serosurveillance (antibody testing) and mosquito pool screening (by RT-qPCR or virus isolation) are currently used to monitor arbovirus transmission, however substantial time lags of seroconversion and/or laborious mosquito identification and RNA extraction steps sacrifice their early warning value. As a consequence, timely vector control responses are compromised. Here, we report on development of a rapid arbovirus detection system whereby adding sucrose to reagents of loop-mediated isothermal amplification with displaced probes (DP-LAMP) elicits infectious mosquitoes to feed directly upon the reagent mix and expectorate viruses into the reagents during feeding. We demonstrate that RNA from pathogenic arboviruses (West Nile and Dengue viruses) transmitted in the infectious mosquito saliva was detectable rapidly (within 45 minutes) without RNA extraction. Sucrose stabilized viral RNA at field temperatures for at least 48 hours, important for transition of this system to practical use. After thermal treatment, the DP-LAMP could be reliably visualized by a simple optical image sensor to distinguish between positive and negative samples based on fluorescence intensity. Field application of this technology could fundamentally change conventional arbovirus surveillance methods by eliminating laborious RNA extraction steps, permitting arbovirus monitoring from additional sites, and substantially reducing time needed to detect circulating pathogens.

Metabarcoding mosquitoes: MinION sequencing of bulk samples gives accurate species profiles for vector surveillance (Culicidae)

Mosquitoes (Family: Culicidae) are dominant vectors of pathogens, and their surveillance has been incorporated into major disease control programs worldwide. However, routine, species-level identification of mosquitoes is often a bottleneck for management, and Next Generation Sequencing (NGS) platforms and DNA metabarcoding can revolutionize this process. MinION nanopore technologies promise on-site sequencing and rapid sample processing rates ideal for time-sensitive biosurveillance. Here, we benchmark the results of DNA metabarcoding on the MinION against the Illumina MiSeq platform, which is known for its higher sequencing accuracy. We used metazoan COI mini-barcode primers to carry out DNA metabarcoding of mosquito bulk samples caught during a real vector survey, then compared the mosquito species profiles recovered on each sequencing platform. We also tested the influence of using different trap lures, storage methods, and pooling different specimen body parts on the number of species recovered. We report that mosquito species-level identifications were highly congruent between MinION and Illumina (93% overlap). We also find that CO2 gas cylinders outperformed biogenic CO2 sources significantly, by two-fold. Notably, we demonstrated the feasibility of detecting zoonotic reservoirs and pathogen signals from mosquito bulk samples. We present the first use of DNA metabarcoding on the MinION for vector surveillance and discuss future applications.

Automated differentiation of mixed populations of free-flying female mosquitoes under semi-field conditions

Great advances in automated identification systems, or ‘smart traps’, that differentiate insect species have been made in recent years, yet demonstrations of field-ready devices under free-flight conditions remain rare. Here, we describe the results of mixed-species identification of female mosquitoes using an advanced optoacoustic smart trap design under free-flying conditions. Point-of-capture classification was assessed using mixed populations of congeneric (Aedes albopictus and Aedes aegypti) and non-congeneric (Ae. aegypti and Anopheles stephensi) container-inhabiting species of medical importance. Culex quinquefasciatus, also common in container habitats, was included as a third species in all assessments. At the aggregate level, mixed collections of non-congeneric species (Ae. aegypti, Cx. quinquefasciatus, and An. stephensi) could be classified at accuracies exceeding 90% (% error = 3.7–7.1%). Conversely, error rates increased when analysing individual replicates (mean % error = 48.6; 95% CI 8.1–68.6) representative of daily trap captures and at the aggregate level when Ae. albopictus was released in the presence of Ae. aegypti and Cx. quinquefasciatus (% error = 7.8–31.2%). These findings highlight the many challenges yet to be overcome but also the potential operational utility of optoacoustic surveillance in low diversity settings typical of urban environments.

Convolutional neural network-based real-time mosquito genus identification using wingbeat frequency: A binary and multiclass classification approach

Global rises in dengue hemorrhagic fever, especially in Asia and Latin America, underscore the necessity for enhanced public health interventions. Aedes spp. mosquitoes are the primary vectors; however, species such as Culex quinquefasciatus pose significant health risks by transmitting diseases such as filariasis, impacting millions of people worldwide. This study introduces a real-time convolutional neural network-based mosquito classification system using wingbeat frequency for identifying various mosquito species, with emphasis on Aedes sp. We proposed and assessed two models: a binary classification and a multiclass system. The binary system exhibited an outstanding accuracy of 91.76% in distinguishing between Aedes aegypti and Culex quinquefasciatus. The multiclass system accurately identified female and male Aedes aegypti and Culex quinquefasciatus with a precision of 87.16%. This innovative approach serves as a potential tool for dengue infection control and a versatile instrument for combating various mosquito-borne illnesses, enhancing vector surveillance for comprehensive disease management.

Tracking arboviruses, their transmission vectors and potential hosts by nanopore sequencing of mosquitoes.

The risk to human health from mosquito-borne viruses such as dengue, chikungunya and yellow fever is increasing due to increased human expansion, deforestation and climate change. To anticipate and predict the spread and transmission of mosquito-borne viruses, a better understanding of the transmission cycle in mosquito populations is needed. We present a pathogen-agnostic combined sequencing protocol for identifying vectors, viral pathogens and their hosts or reservoirs using portable Oxford Nanopore sequencing. Using mosquitoes collected in São Paulo, Brazil, we extracted RNA for virus identification and DNA for blood meal and mosquito identification. Mosquitoes and blood meals were identified by comparing cytochrome c oxidase I (COI) sequences against a curated Barcode of Life Data System (BOLD). Viruses were identified using the SMART-9N protocol, which allows amplified DNA to be prepared with native barcoding for nanopore sequencing. Kraken 2 was employed to detect viral pathogens and Minimap2 and BOLD identified the contents of the blood meal. Due to the high similarity of some species, mosquito identification was conducted using blast after generation of consensus COI sequences using RACON polishing. This protocol can simultaneously uncover viral diversity, mosquito species and mosquito feeding habits. It also has the potential to increase understanding of mosquito genetic diversity and transmission dynamics of zoonotic mosquito-borne viruses.

Cladogram construction of mosquitos based on morphological characteristics

The mosquito identification process is very important regarding their role as vectors in many arthropod-borne diseases. The mosquito's identification based on morphological characteristics is very complex. The constraint in the identification process is the unavailability of adequate human resources. This article presents an effort to identify mosquitoes more easily based on the specific characteristics of each taxon. The aim of making a cladogram is to make it easier to identify mosquitoes in family, genus, and species taxa, especially in the genus Anopheles. Cladograms were constructed based on morphological characters on wings and their pale scale pattern, scutellum, proboscis, and palpi. The study revealed that based on morphological characters, four cladograms were constructed based on the number of wings, the number of pale scales four or more, and also less than three pale scales, that were for Genus Anopheles, Subgenus Anopheles, and two cladograms for Subgenus Cellia. The specific morphological characters in mosquitos can be used to construct the cladogram for easier identification.