Anopheles coluzzii (Wild type Ngousso strain; genotype: Orco-QF2 , QUAS-GCaMP6f), Aedes aegypti (Wild type LVPib12 strain), and Culex quinquefasciatus mosquitoes (Wild type Johannesburg strain) were raised in a climate chamber maintained at 26-28°C, 70-80% RH and L14:D10 cycle. Eggs were hatched in deionized water, and larvae fed on fish food (TetraMin®), added daily. Three days after hatching, larvae were counted and kept at a density of ~150 larvae/L of water. Cotton rolls soaked with a sugar solution (10% w/vol) were provided to feed adult mosquitoes. Colony female mosquitoes were blood-fed on mice according to a protocol approved by the Johns Hopkins University Animal Care and Use Committee. For all experiments, non blood-fed female mosquitoes (3-10 days old) that were allowed to mate freely were used.
Odorants were purchased at the highest purity available. 1-octen-3-ol (SAFC, product # W280518), and benzaldehyde (Aldrich, product # 418099) were used undiluted or diluted in paraffin oil (Sigma-Aldrich, product# 18512) to 0.1%, 1%, and 10%. IR3535 (EMD Chemicals, product# 111887), picaridin (Cayman Chemical, product# 16458), and eugenol (Aldrich, product# E51791) were used undiluted. Lemongrass oil (SAFC, product# W262404), p-menthane-3,8-diol (BOC Sciences, 80%, catalog# B0005-092293), and DEET (Sigma Aldrich, product # 36542), were used undiluted, or diluted in paraffin oil to 30%.
Close proximity response assay
Female mosquitoes were tested individually (a total of 30 mosquitoes for each experiment). A mosquito was transferred to a cage (BugDorm, 30 x 30 x 30 cm) and given enough time (≥5 minutes) to come to rest on one of the cage mesh walls (Fig. 1b). After 30 seconds at rest, the mosquito was then approached from outside the cage by a 1000 μl pipette tip (Denville) containing a piece of filter paper (1x2 cm) soaked with 20 µl of the test odorant. The pipette tip, held by a gloved hand, was rested on the outside of the cage wall so that the mosquito was at a 0.5 cm distance from the filter paper. The mosquito was observed for 30 seconds and the time at which it flew away was scored. The sequence of the odorants was randomized every time, and the mosquito was given ≥2 minutes between odorants. A mosquito that flew off in response to an odorant was allowed to land and rest for ≥2 minutes before the next odorant was used.
Analysis of close proximity response assay
A “Kaplan-Meier Survival Estimates” was used to summarize the time that all 30 tested mosquitoes took to fly in response to odorants. A Cox Proportional Hazard Model was then used to assess significant differences in response time, which also considered the number of previous odorant exposures. The plot and analysis were performed using “survival” and “survminer” packages in R .
Only female mosquitoes were used. A mosquito was immobilized on ice for 1 min, inserted into a pipette tip, and pushed so that only its antennae extended outside the pipette tip. The pipette tip was then mounted onto a glass slide using modeling clay. The antenna was placed forward and held down on a glass cover slip using two pulled glass capillary tubes (Harvard Apparatus, 1 OD x 0.5 ID x 100 L mm). One tube was used to hold down the 3rd-4th antennal segment, and the other glass tube was used to hold down the 12th-13th segment (the most distal segments). All recordings focused on one antennal segment (11th antennal segment). Previous recordings found that the responses of this one segment (11th segment) were representative of olfactory responses from the other segments .
Imaging was performed through a 50x objective (LD EC Epiplan-Neofluar 50x/0.55 DIC) mounted on a Zeiss Axio Examiner D1 microscope. A Zeiss Illuminator HXP 200C light source and an eGFP filter cube (FL Filter Set 38 HE GFP shift free) were used for fluorescence.
An EMCCD camera (Andor iXon Ultra, Oxford Instruments) using Andor Solis software (Oxford Instruments) was used to record videos of 20 seconds at 512x512 pixel resolution. The exposure time was 200 ms (5 Hz).
Odorant preparation and delivery
For testing human-odorants, 20 µl of the odorant solution was pipetted onto a piece of filter paper (1x2 cm) that was placed in a Pasteur pipette (Fisher Scientific). For single repellents, 10 µl of the repellent (at 60%) was pipetted on the same filter paper with 10 µl paraffin oil. To prepare repellent mixtures, 10 µl of each repellent (at 60%) were pipetted on the same filter paper to reach the desired final concentration when mixed (30% of each repellent). The Pasteur pipette was inserted into a hole in a plastic serological pipette (Denville Scientific Inc, 10ml pipette) carrying a continuous stream of purified air (8.3 ml/s) directed towards the antenna (Fig. 3b). A stimulus controller (Syntech) was used to divert a 1 s pulse of charcoal-filtered air (5 ml/s) into the Pasteur pipette 10 seconds after the beginning of each recording. The sequence of odorants was randomized for each set of experiments, and new Pasteur pipettes were prepared for each recording session.
Analysis of calcium imaging recordings
To generate heatmap ΔF images, we used Fiji software  with a custom-built macro. This macro uses the "Image stabilizer" plug-in to correct for movement, followed by the "Z project" function to calculate the mean baseline fluorescence. The mean baseline fluorescence was represented by the first 9 seconds of recording, just before stimulus delivery. The "Image calculator" function was then used to subtract the mean baseline fluorescence from the maximum fluorescence frame after odorant delivery (this image was manually chosen). This calculated ΔF image was then used to produce the heatmaps in the figures.