Phenotypic variation in populations of the mosquito vector, Aedes aegypti, and implications for predicting the effects of temperature and climate change on dengue transmission
Data files
Nov 10, 2025 version files 2.95 MB
-
Dennington_etal._Data_MX_TPC_.zip
2.94 MB
-
README.md
6.61 KB
Abstract
There is concern that increases in temperature due to climate change could lead to shifts in the dynamics and distribution of mosquito vectors. Many current models assume there are 'average' thermal performance curves for a given vector species transmission. However, this ‘one-size-fits-all’ assumption ignores the potential for local adaptation to create population-specific differences in thermal performance. In this study, we explored thermal performance of five independent field populations of Ae. aegypti from Mexico, together with a standard laboratory strain. We reared these six populations at temperatures between 13°C and 37°C to generate thermal performance curves for a suite of life-history traits. Composite models integrating these traits revealed the effects of temperature on population growth rates and dengue virus transmission potential. The results provide strong evidence for the potential for local adaptation in Ae. aegypti populations, challenging applicability of ‘one-size-fits-all’ thermal performance models to assess climate impact on mosquito-borne diseases.
https://doi.org/10.5061/dryad.sxksn03bm
We explored standing variation in the mosquito vector species Aedes aegypti through life history experiments. We used a Bayesian approach to analyze these data and produce population growth and temperature dependent R0 models.
Description of the data and file structure
There are three types of files found in this set: Excel/csv files containing raw data, R code, and text files (containing information for Bayesian models). There is one set of data containing life history data for experimental mosquitoes. Life history thermal performance curves correspond to models using a Bayesian approach. Models for mosquito development rate, egg-to-adult survival, fecundity, juvenile mortality rate, and biting rate models are all contained in one R code, and separately are models for adult survival, fitness, and R0 models. Each of these traits requires a .bug/.txt file which contains the model.
Provided in this code is an example in which the models for one population are described in full. These steps can be followed for the models for each of the six populations.
Vector competence models and methods are adapted from Mordecai et al., 2017 and Johnson et al., 2015 and raw data can be found through Vectorbyte.org. Here, I've provided the posterior samples used in transmission models for vector competence (DENV_ParameterFits_b and DENV_ParameterFits_c) and Pathogen development rate (called DENV_ParameterFits_PDR).
The raw data are:
"MX_development.csv":
Includes data from life history measurements including columns Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Temp (Temperature at which the groups were reared, measured in °C), Rep (replicate measurements of each selected line), Pupae (number of pupae formed), Adults (number of adults eclosed), Larv*Surv *(proportion of larvae that survived to adult from starting 200 larvae), Pupal death (number of pupae that formed but did not eclose), Time to pupa (days to pupation), Time to adult (days to adult), dev rate p (pupation rate= 1/ time to pupa), dev rate_a (mosquito development rate= 1/time to adult), number attempted (total number of starting mosquitoes), and juvenileMU (juvenile mortality rate- approximation using the -(natural log of Larval survival)/ Time to adult).
"MX_fecundity.csv":
Includes data from life history measures for fecundity including Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Temp (Temperature at which the groups were reared, measured in °C), Rep (replicate measurements of each selected line), number (Individual within each experimental group), eggs (number of eggs for the first gonotrophic cycle), larvae (number of larvae that hatched), hatch_rate (number of larvae/number of eggs), date_fed (date of the first blood feed), date_first_lay (date of the first time eggs were layed), and cycle_length (amount of time in days between first blood feed to first egg lay).
"MX_adultsurvraw.csv":
Includes data from life history measures for adult survival in a summary form. Columns are Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Temp (Temperature at which the groups were reared, measured in °C), Rep (replicate measurements of each selected line), Date (Date the individual died), Day (Day of the experiment), M (number of dead males), F (number of dead females), DeadHouse1 (number of dead females in first enclosure), DeadHouse2 (number of dead females in second enclosure), DeadHouseFalcon (number of dead females in third enclosure ), ALL_dead_F (total number of dead females from all three enclosures). We have also included the file for "MXAdultSurvivalCleaned.Rsave", which is the formatted version of the adult survival raw data needed to run the model in "MexicoTPC_Code.R".
The folders include:
“MX TPC data” which are the data used in the models including raw data and summarized data for vector competence. “Models” which includes the JAGS models used to run the R code.
The summarized data includes:
"MX_SampleSizes"- Includes the summary of actual sample sizes represented across population, temperature and replicate. Includes three data sheets: "Larval Development Sample Sizes", "Adult Survival Sample Size", and "Fecundity Sample Size". The Larval development sample sizes file includes Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Rep (replicate measurements of each selected line), Temp (Temperature at which the groups were reared), Pupae (Number of mosquitoes that pupated out of 200), and Adults (Number of mosquitoes that emerged from the number of pupae). "Adult Survival Sample Size" include the columns Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Rep (replicate measurements of each selected line), Temp (Temperature at which the groups were reared), Total.Dead.Adults( Number of mosquitoes that were counted as dead before censoring), Number.Adults.Eclosed (Number of mosquitoes that emerged from the number of pupae) and Day (the day at which survival was measured until from first instar). "Fec Sample Size" includes columns Population (Acapulco, Cabo, Monterrey, Ciudad Juarez, Jojutla and Lab), Temp (Temperature at which the groups were reared), Rep (replicate measurements of each selected line), Individual count (the number of individuals measured for fecundity), Individual or group? (whether fecundity was measured as individuals or as a group of 10 or less).
The files including the code to produce all models are:
"Dennington_MexicoTPC_Code.R"- Includes the code for biting rate, mosquito development rate, egg to adult survival probability, fecundity, juvenile mortality rate, adult survival, temperature dependent transmission and population growth. Note that these require the models "JAGS_model_briere.bug", "JAGS_Modelexample.bug", ""surv_gengamma_temp.txt", "mcmc_utils.R", and "temp_functions.R". The temperature dependent transmission model requires the "DENV_ParameterFits" posterior samples (3 total), which are from Mordecai et al., 2017 and Johnson et al., 2015. In order to produce the models for each population, run the code for each population and save the posterior samples. After, run the code to produce the composite fitness models.
Code/Software
This code was run on R/Rstudio Version 2023.06.1.
Methods used in the current study closely follow those outlined in Dennington et al. 2024.
Mosquito Collection
Aedes aegypti mosquitoes were collected from the field in five different locations in Mexico (Cabo San Lucas, Acapulco, Monterrey, Ciudad Juárez, and Jojutla) using ovitraps. Populations from Mexico were founded with at least 100 females and to remove the influence of maternal effects, mosquitoes were reared in the lab for at least one generation in standard laboratory conditions (27°C, 80% humidity, 12:12hr photoperiod) prior to experimentation. Two populations, Jojutla and Ciudad Juárez, were reared for an additional generation to ensure a large enough population for subsequent experiments. The field locations were chosen to capture a gradient of the climate and landscape. The field populations were compared to a standard laboratory population (Rockefeller strain) that were maintained at Penn State University under standard insectary conditions over many years.
Experiments to Generate Temperature-Dependent Data
Mosquito life-history traits including egg-to-adult survival, mosquito development rate, mean adult survival, fecundity, and biting rate were measured in mosquitoes reared at 13°C, 15°C, 19°C, 23°C, 25°C, 27°C, 29°C, 31°C, 33°C, 35°C, 37°C, each ± 0.2°C and 80 ± 10% relative humidity in environmentally controlled incubators. These life-history measurements were replicated three times at each temperature for each population. We began with eggs from the five field populations and one laboratory line were hatched at 27°C for 24 hours. Then, 200 first instar larvae were put into 1.89 L containers with 1 L of deionized water and 0.20 mg of larvae bovine liver powder (MP Biomedicals) and each of the three replicates were placed in the incubator at their respective temperatures. We fed larvae 0.20 mg of liver powder per larvae every other day until pupation, but once pupation began we scaled their food to the remaining number of larvae. Pupae, both living and dead, were removed and counted on the day of pupation and placed in a small cup (30 mL) with water from their original environment to allow for eclosion. Pupae were then added to a small cage (17.5 cm3) with continuous** access to 10% sugar solution (dextrose anhydrous and deionized water). We counted the number of adults that eclosed every day. After 95% of surviving females emerged, we blood-fed females after 3-5 days. We used blood from de-identified human donors (BioIVT, Corp.) and so IRB approval and human subjects’ approval was not needed. We immediately counted the total number of blood-fed females and placed up to 10 individual females into separate containers (50 mL polypropylene centrifuge tubes) that were lined with filter paper and 7 mL deionized water to measure individual fecundity. We also placed up to 20 females into two small cages (10 in each) with a small filter paper for egg laying to monitor adult survival. We recorded the day that females in individual containers first laid eggs, which we used for fecundity measures and to approximate the biting rate (1/gonotrophic cycle length), after which we removed them from their containers and placed them into the group cages. We extracted the water from the containers to let the filter paper dry in their respective incubators and then we counted the number of eggs from individual mosquitoes. For the course of the experiment, we offered each cage of females a blood meal every 4 days, and counted the number of adults that died every day. We censored this experiment 4 weeks after the first egg lay at each temperature.