Age grading An. gambiae and An. arabiensis using near infrared spectra and artificial neural networks.
ABSTRACT: BACKGROUND:Near infrared spectroscopy (NIRS) is currently complementing techniques to age-grade mosquitoes. NIRS classifies lab-reared and semi-field raised mosquitoes into < or ? 7 days old with an average accuracy of 80%, achieved by training a regression model using partial least squares (PLS) and interpreted as a binary classifier. METHODS AND FINDINGS:We explore whether using an artificial neural network (ANN) analysis instead of PLS regression improves the current accuracy of NIRS models for age-grading malaria transmitting mosquitoes. We also explore if directly training a binary classifier instead of training a regression model and interpreting it as a binary classifier improves the accuracy. A total of 786 and 870 NIR spectra collected from laboratory reared An. gambiae and An. arabiensis, respectively, were used and pre-processed according to previously published protocols. The ANN regression model scored root mean squared error (RMSE) of 1.6 ± 0.2 for An. gambiae and 2.8 ± 0.2 for An. arabiensis; whereas the PLS regression model scored RMSE of 3.7 ± 0.2 for An. gambiae, and 4.5 ± 0.1 for An. arabiensis. When we interpreted regression models as binary classifiers, the accuracy of the ANN regression model was 93.7 ± 1.0% for An. gambiae, and 90.2 ± 1.7% for An. arabiensis; while PLS regression model scored the accuracy of 83.9 ± 2.3% for An. gambiae, and 80.3 ± 2.1% for An. arabiensis. We also find that a directly trained binary classifier yields higher age estimation accuracy than a regression model interpreted as a binary classifier. A directly trained ANN binary classifier scored an accuracy of 99.4 ± 1.0 for An. gambiae and 99.0 ± 0.6% for An. arabiensis; while a directly trained PLS binary classifier scored 93.6 ± 1.2% for An. gambiae and 88.7 ± 1.1% for An. arabiensis. We further tested the reproducibility of these results on different independent mosquito datasets. ANNs scored higher estimation accuracies than when the same age models are trained using PLS. Regardless of the model architecture, directly trained binary classifiers scored higher accuracies on classifying age of mosquitoes than regression models translated as binary classifiers. CONCLUSION:We recommend training models to estimate age of An. arabiensis and An. gambiae using ANN model architectures (especially for datasets with at least 70 mosquitoes per age group) and direct training of binary classifier instead of training a regression model and interpreting it as a binary classifier.
Project description:After mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases in humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which are currently used to determine the parity status of mosquitoes, are very tedious and limited to few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes. In this study, we train an artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets: An. arabiensis collected from Minepa, Tanzania (Minepa-ARA); An. gambiae s.s collected from Muleba, Tanzania (Muleba-GA); An. gambiae s.s collected from Burkina Faso (Burkina-GA); and An.gambiae s.s from Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train the ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9±2.8 (N = 274), 68.7±4.8 (N = 43), 80.3±2.0 (N = 48), and 75.7±2.5 (N = 91), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1±2.2% (N = 274), 89.8 ± 1.7% (N = 43), 93.3±1.2% (N = 48), and 92.7±1.8% (N = 91) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively. These results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate the parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.
Project description:Understanding the age-structure of mosquito populations, especially malaria vectors such as Anopheles gambiae, is important for assessing the risk of infectious mosquitoes, and how vector control interventions may impact this risk. The use of near-infrared spectroscopy (NIRS) for age-grading has been demonstrated previously on laboratory and semi-field mosquitoes, but to date has not been utilized on wild-caught mosquitoes whose age is externally validated via parity status or parasite infection stage. In this study, we developed regression and classification models using NIRS on datasets of wild An. gambiae (s.l.) reared from larvae collected from the field in Burkina Faso, and two laboratory strains. We compared the accuracy of these models for predicting the ages of wild-caught mosquitoes that had been scored for their parity status as well as for positivity for Plasmodium sporozoites.Regression models utilizing variable selection increased predictive accuracy over the more common full-spectrum partial least squares (PLS) approach for cross-validation of the datasets, validation, and independent test sets. Models produced from datasets that included the greatest range of mosquito samples (i.e. different sampling locations and times) had the highest predictive accuracy on independent testing sets, though overall accuracy on these samples was low. For classification, we found that intramodel accuracy ranged between 73.5-97.0% for grouping of mosquitoes into "early" and "late" age classes, with the highest prediction accuracy found in laboratory colonized mosquitoes. However, this accuracy was decreased on test sets, with the highest classification of an independent set of wild-caught larvae reared to set ages being 69.6%.Variation in NIRS data, likely from dietary, genetic, and other factors limits the accuracy of this technique with wild-caught mosquitoes. Alternative algorithms may help improve prediction accuracy, but care should be taken to either maximize variety in models or minimize confounders.
Project description:BACKGROUND: The dramatic success of insecticide treated nets (ITNs) and long-lasting insecticidal nets (LLINs) in African countries has been countered by the rapid development of pyrethroid resistance in vector mosquitoes over the past decade. One advantage of the use of pyrethroids in ITNs is their excito-repellency. Use of the excito-repellency of pyrethroids might be biorational, since such repellency will not induce or delay the development of any physiological resistance. However, little is known about the relationship between the mode of insecticide resistance and excito-repellency in pyrethroid-resistant mosquitoes. METHODS: Differences in the reactions of 3 major malaria vectors in western Kenya to pyrethroids were compared in laboratory tests. Adult susceptibility tests were performed using World Health Organization (WHO) test tube kits for F1 progenies of field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s., and laboratory colonies of An. gambiae s.s. and An. arabiensis. The contact repellency to pyrethroids or permethrin-impregnated LLINs (Olyset® Nets) was evaluated with a simple choice test modified by WHO test tubes and with the test modified by the WHO cone bioassay test. RESULTS: Field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s. showed high resistance to both permethrin and deltamethrin. The allelic frequency of the point mutation in the voltage-gated sodium channel (L1014S) in An. gambiae s.s. was 99.3-100%, while no point mutations were detected in the other 2 species. The frequency of takeoffs from the pyrethroid-treated surface and the flying times without contacting the surface increased significantly in pyrethroid-susceptible An. gambiae s.s. and An. arabiensis colonies and wild An. arabiensis and An. funestus s.s. colonies, while there was no significant increase in the frequency of takeoffs or flying time in the An. gambiae s.s. wild colony. CONCLUSION: A different repellent reaction was observed in the field-collected An. gambiae s.s. than in An. arabiensis and An. funestus s.s. It might be that resistant mosquitoes governed by knockdown resistance (kdr) loose repellency to pyrethroids, whereas those lacking kdr maintain high repellency irrespective of their possessing metabolic resistance factors to pyrethroids. Further genetic evaluation is required for the demonstration of the above hypothesis.
Project description:AgDNV is a powerful gene transduction tool and potential biological control agent for Anopheles mosquitoes. Using a GFP reporter virus system, we investigated AgDNV host range specificity in four arthropod cell lines (derived from An. gambiae, Aedes albopictus and Drosophila melanogaster) and six mosquito species from 3 genera (An. gambiae, An. arabiensis, An. stephensi, Ae. albopictus, Ae. aegypti and Culex tarsalis). In vitro, efficient viral invasion, replication and GFP expression was only observed in MOS55 An. gambiae cells. In vivo, high levels of GFP were observed in An. gambiae mosquitoes. Intermediate levels of GFP were observed in the closely related species An. arabiensis. Low levels of GFP were observed in An. stephensi, Ae. albopictus, Ae. aegypti and Cx. tarsalis. These results suggest that AgDNV is a specific gene transduction tool for members of the An. gambiae species complex, and could be potentially developed into a biocontrol agent with minimal off-target effects.
Project description:Anopheles gambiae s.l. are important malaria vectors, but little is known about their genomic variation in the wild. Here, we present inter- and intraspecies analysis of genome-wide RADseq data, in three Anopheles gambiae s.l. species collected from East Africa. The mosquitoes fall into three genotypic clusters representing described species (A. gambiae, A. arabiensis, and A. merus) with no evidence of cryptic breeding units. Anopheles merus is the most divergent of the three species, supporting a recent new phylogeny based on chromosomal inversions. Even though the species clusters are well separated, there is extensive shared polymorphism, particularly between A. gambiae and A. arabiensis. Divergence between A. gambiae and A. arabiensis does not vary across the autosomes but is higher in X-linked inversions than elsewhere on X or on the autosomes, consistent with the suggestion that this inversion (or a gene within it) is important in reproductive isolation between the species. The 2La/2L+(a) inversion shows no more evidence of introgression between A. gambiae and A. arabiensis than the rest of the autosomes. Population differentiation within A. gambiae and A. arabiensis is weak over approximately 190-270 km, implying no strong barriers to dispersal. Analysis of Tajima's D and the allele frequency spectrum is consistent with modest population increases in A. arabiensis and A. merus, but a more complex demographic history of expansion followed by contraction in A. gambiae. Although they are less than 200 km apart, the two A. gambiae populations show evidence of different demographic histories.
Project description:BACKGROUND:A three-year longitudinal study was conducted in four sentinel sites from different ecological settings in Burkina Faso, between 2008 and 2010 to identify longitudinal changes in insecticide resistance within Anopheles gambiae complex species based on larval collection. During this study, adult mosquitoes were also collected indoor and outdoor using several methods of collection. The present study reports the diversity of malaria vectors and the 1014F-genotype from this adult collection and investigates the association between this 1014F-genotype and sporozoite rate. METHODS:Adult mosquitoes were collected from July to August (corresponding to the start of rainy season) and October to November (corresponding to the end of rainy season) over 3 years (2008-2010) at four sites across the country, using pyrethrum spray catches (PSC), exit traps and pit shelters. Anopheles gambiae complex mosquitoes were identified to species and genotyped for the L1014F kdr mutation by PCR using genomic DNA. The circumsporozoite antigen of Plasmodium falciparum was detected in mosquitoes using sandwich ELISA. RESULTS:Overall 9212 anopheline mosquitoes were collected during the study period. Of those, 6767 mosquitoes were identified as Anopheles gambiae sensu lato (s.l.). Anopheles arabiensis, Anopheles coluzzii, Anopheles gambiae and or Anopheles funestus were incriminated as vectors of P. falciparum in the study area with an average sporozoite rate of 5%, (95% CI 4.14-5.99%). The kdr1014F-genotype frequencies were 11.44% (95% CI 2.5-39.85%), 19.2% (95% CI 4.53-53.73%) and 89.9 (95% CI 63.14-97.45%), respectively for An. arabiensis, An. coluzzii and An. gambiae. The proportion of the 1014F-genotype varied between sporozoite-infected and uninfected An. gambiae s.l. group. There was no significant difference in the 1014F-genotype frequency between infected and uninfected mosquitoes. CONCLUSION:The current study shows the diversity of malaria vectors and significant interaction between species composition and kdr1014F-genotype in An. gambiae complex mosquitoes from Burkina Faso. In this study, no associations were found between the 1014F-genotype and P. falciparum infection in the major malaria vector An. gambiae s.l.
Project description:Heterochromatin is identified as a potential factor driving diversification of species. To understand the magnitude of heterochromatin variation within the Anopheles gambiae complex of malaria mosquitoes, we analyzed metaphase chromosomes in An. arabiensis, An. coluzzii, An. gambiae, An. merus, and An. quadriannulatus. Using fluorescence in situ hybridization (FISH) with ribosomal DNA (rDNA), a highly repetitive fraction of DNA, and heterochromatic Bacterial Artificial Chromosome (BAC) clones, we established the correspondence of pericentric heterochromatin between the metaphase and polytene X chromosomes of An. gambiae. We then developed chromosome idiograms and demonstrated that the X chromosomes exhibit qualitative differences in their pattern of heterochromatic bands and position of satellite DNA (satDNA) repeats among the sibling species with postzygotic isolation, An. arabiensis, An. merus, An. quadriannulatus, and An. coluzzii or An. gambiae. The identified differences in the size and structure of the X chromosome heterochromatin point to a possible role of repetitive DNA in speciation of mosquitoes. We found that An. coluzzii and An. gambiae, incipient species with prezygotic isolation, share variations in the relative positions of the satDNA repeats and the proximal heterochromatin band on the X chromosomes. This previously unknown genetic polymorphism in malaria mosquitoes may be caused by a differential amplification of DNA repeats or an inversion in the sex chromosome heterochromatin.
Project description:Mosquito nets containing synergists designed to overcome metabolic resistance mechanisms in vectors have been developed. These may enhance excitability in the mosquitoes and affect how they respond to CDC light-traps. Investigating the behaviour of vectors of disease in relation to novel mosquito nets is, therefore, essential for the design of sampling and surveillance systems.In an initial experiment in Muleba, Tanzania, nine bedrooms from three housing clusters were sampled. CDC light-traps were operated indoors next to occupied untreated nets (UTN), Olyset® long lasting insecticidal net (LLIN) and Olyset Plus® LLIN containing piperonyl butoxide (PBO) synergist. Nets were rotated daily between the nine rooms over nine nights. A further series of experiments using the nets on alternate nights in a single room was undertaken during the short rains. Anopheles gambiae s.l. were collected in CDC light-traps, a window-trap and Furvela tent-trap. Anopheles gambiae s.l. were identified to species by polymerase chain reaction (PCR).In the initial experiment 97.7% of the 310 An. gambiae s.l. were An. gambiae s.s., the remainder being Anopheles arabiensis. The number of mosquitoes collected from 81 light-trap collections was greater in the presence of an Olyset [density rate ratio 1.81, 95% CI (1.22-2.67), p = 0.003] relative to an UTN. In a second experiment, in the wet season 84% of the 180 An. gambiae s.l. identified were An. arabiensis. The number of An. gambiae s.l. collected from a light-trap compared to a tent-trap was significantly higher when an Olyset Plus net was used compared to an UTN. Survival of the mosquitoes in the window trap was not reduced by the use of an Olyset Plus net in the bedroom relative to an Olyset net.Mosquitoes entering bedrooms, even those susceptible to pyrethroids, were not killed by contact with an Olyset Plus LLIN. The enhanced numbers of An. gambiae or An. arabiensis collected in light-traps when a treated net is used requires further experimentation and may be because of a heightened escape reaction on the part of the mosquito.
Project description:The voltage gated sodium channel mutation Vgsc-1014S (kdr-east) was first reported in Kenya in 2000 and has since been observed to occur at high frequencies in the local Anopheles gambiae s.s.The mutation Vgsc-1014F has never been reported from An. gambiae Complex complex mosquitoes in Kenya.Molecularly confirmed An. gambiae s.s. (hereafter An. gambiae) and An. arabiensis collected from 4 different parts of western Kenya were genotyped for kdr from 2011 to 2013. Vgsc-1014F was observed to have emerged, apparently, simultaneously in both An. gambiae and An. arabiensis in 2012. A portion of the samples were submitted for sequencing in order to confirm the Vgsc-1014F genotyping results. The resulting sequence data were deposited in GenBank (Accession numbers: KR867642-KR867651, KT758295-KT758303). A single Vgsc-1014F haplotype was observed suggesting, a common origin in both species.This is the first report of Vgsc-1014F in Kenya. Based on our samples, the mutation is present in low frequencies in both An. gambiae and An. arabiensis. It is important that we start monitoring relative frequencies of the two kdr genes so that we can determine their relative importance in an area of high insecticide treated net ownership.
Project description:Mosquitoes belonging to Anopheles gambiae species complex are the main malaria vector in Mauritania but data on their vector capacities, feeding habits and insecticide susceptibility are still scanty. The objectives of this study were to fill this gap.Adult Anopheles spp. mosquitoes were collected using pyrethrum spray catch method from two ecological zones of Mauritania: Nouakchott (Saharan zone) and Hodh Elgharbi region (Sahelian zone). Circumsporozoite proteins (CSP) for P. falciparum, P. vivax VK210 and P. vivax VK247 were detected by enzyme-linked immunosorbent assay (ELISA) from the female anopheline mosquitoes. To confirm CSP-ELISA results, polymerase chain reaction (PCR) was also performed. Blood meal identification was performed in all engorged females by partial sequencing of the mitochondrial cytochrome b gene. Molecular assessments of pyrethroid knockdown resistance (kdr) and insensitive acetylcholinesterase resistance (ace-1) were conducted.In Nouakchott, the only species of Anopheles identified during the survey was Anopheles arabiensis (356 specimens). In Hodh Elgharbi, 1016 specimens of Anopheles were collected, including 578 (56.9%) Anopheles rufipes, 410 (40.35%) An. arabiensis, 20 (1.96%) An. gambiae, 5 (0.5%) An. pharoensis and 3 (0.3 %) An. funestus. Three of 186 female An. arabiensis collected in Nouakchott and tested by ELISA were found positive for Plasmodium vivax VK210, corresponding to a sporozoite rate of 1.6%; however PCR confirmed infection by P. vivax sporozoite in only one of these. In Hodh Elgharbi, no mosquito was found positive for Plasmodium spp. infection. There was a statistically significant difference in the percentage of human blood-fed Anopheles spp. between Nouakchott (58.7%, 47 of 80 blood-engorged An. arabiensis females) and Hodh Elgharbi (11.1%, 2 of 18 blood-engorged mosquitoes). Analysis of the kdr polymorphisms showed 48.2% (70/145) of East African kdr mutation (L1014S) in Nouakchott compared to 10% (4/40) in Hodh Elgharbi region (P?<?0.001). Nevertheless, West African kdr mutation (L1014F) was found only in An. gambiae populations (4/40, 10%) from Hodh Elgharbi region. No ace-1 mutation was found in mosquito specimens from the two study zones.Overall, this study confirmed the autochthonous P. vivax malaria transmission in Nouakchott, involving An. arabiensis as the main vector. It also described for the first time the absence of ace-1 mutation, the co-occurrence of both West and East African kdr mutation in An. gambiae in Mauritania, and highlighted the regional variations in the prevalence and type of kdr mutations.