An ensemble learning approach to map the genetic connectivity of the parasitoid Stethynium empoasca (Hymenoptera: Mymaridae) and identify the key influencing environmental and landscape factors

The effect of landscape patterns and environmental factors on the population structure genetic diversity organisms is well-documented. However, this still unclear in case Mymaridae parasitoids. Despite recent advances machine learning methods for genetics, ensemble needs further investigation. Here, we evaluated performance different boosting algorithms analyzed effects variations tea green leafhopper parasitoid Stethynium empoasca (Hymenoptera: Mymaridae). S. populations showed a distinct pattern isolation by distance. minimum temperature coldest month, annual precipitation, coverage evergreen/deciduous needleleaf trees per 1 km 2 , precipitation warmest quarter were identified as dominant affecting divergence populations. Notably, compared to previous studies, our model an unprecedented accuracy ( r = 0.87) prediction differentiation. These findings not only demonstrated how shaped genetics but also provided essential basis developing conservation strategies biocontrol agent. In broader sense, study importance efficiency genetics.