Data-driven and interpretable machine-learning modeling to explore the fine-scale environmental determinants of malaria vectors biting rates in rural Burkina Faso

Abstract

Background Improving the knowledge and understanding of the environmental determinants of malaria vector abundance at fine spatiotemporal scales is essential to design locally tailored vector control intervention. This work is aimed at exploring the environmental tenets of human-biting activity in the main malaria vectors (Anopheles gambiae s.s., Anopheles coluzzii and Anopheles funestus) in the health district of Diébougou, rural Burkina Faso. Methods Anopheles human-biting activity was monitored in 27 villages during 15 months (in 2017–2018), and environmental variables (meteorological and landscape) were extracted from high-resolution satellite imagery. A two-step data-driven modeling study was then carried out. Correlation coefficients between the biting rates of each vector species and the environmental variables taken at various temporal lags and spatial distances from the biting events were first calculated. Then, multivariate machine-learning models were generated and interpreted to (i) pinpoint primary and secondary environmental drivers of variation in the biting rates of each species and (ii) identify complex associations between the environmental conditions and the biting rates. Results Meteorological and landscape variables were often significantly correlated with the vectors’ biting rates. Many nonlinear associations and thresholds were unveiled by the multivariate models, for both meteorological and landscape variables. From these results, several aspects of the bio-ecology of the main malaria vectors were identified or hypothesized for the Diébougou area, including breeding site typologies, development and survival rates in relation to weather, flight ranges from breeding sites and dispersal related to landscape openness. Conclusions Using high-resolution data in an interpretable machine-learning modeling framework proved to be an efficient way to enhance the knowledge of the complex links between the environment and the malaria vectors at a local scale. More broadly, the emerging field of interpretable machine learning has significant potential to help improve our understanding of the complex processes leading to malaria transmission, and to aid in developing operational tools to support the fight against the disease (e.g. vector control intervention plans, seasonal maps of predicted biting rates, early warning systems). Graphical abstract

Physiological and behavioural resistance of malaria vectors in rural West-Africa : a data mining study to address their fine-scale spatiotemporal heterogeneity, drivers, and predictability

“…Rainfall and temperatures were averaged in a 2 km buffer zone around each HLC collection point to create meteorological variables on both the day of collection and the month preceding collection. Detailed descriptions of the methods used to collect and process these data are provided in Taconet et al (2021).…”

“…In addition, the geographical location of the households was recorded, and used to derive two indices : the degree of clustering of the households in each village, and the distance from each collection point to the edge of the village. Detailed descriptions of the methods used to collect, generate and process these data are provided in Taconet et al (2021).…”

“…First, we found many statistically significant associations between the likelihood of collecting a behaviourally resistant Anopheles and the meteorological conditions during the month preceding collection. This could mean that there might exist a fitness cost to behavioural resistance, associated with (i) maternal or paternal effects or (ii) development or survival at the larval stage (see Taconet et al (2021)), that support (at least for the option (ii) ) the hypothesis of genetic bases for some host-seeking behavioural phenotypes. Second, the exophagy rates of the main malaria vectors in both areas (on average, 41% in the Diébougou area and 56% in the Korhogo area) were substantially higher than those, overall, historically reported for these species ; in general in Africa (Huho et al 2013;Sinka et al 2010) and especially in Burkina Faso (Sherrard-Smith et al 2019).…”

“…At small spatiotemporal scales, this calls for different strategies for respectively vector control (interventions aiming at reducing the human-vector contact) and resistance management (interventions aiming at reducing the development of physiological or behavioural resistance) at such spatiotemporal scales. While vector control plans should be very locally-tailored (species-, season-, and village-specific) (Taconet et al 2021), resistance management strategies would probably not gain much in being adapted to the season or village within our areas. In other words, while resistance management plans are undoubtedly urgently needed, there is no compelling evidence -in the current state of the knowledge -that they should be tailored at very fine scales (village, season).…”

“…CC-BY-NC 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made (Taconet et al 2021).…”