Mosquito chronological age determination using mid-infrared spectroscopy and chemometrics

WCU Author/Contributor (non-WCU co-authors, if there are any, appear on document)
Bradley Forrest Guilliams (Creator)
Institution
Western Carolina University (WCU )
Web Site: http://library.wcu.edu/
Advisor
Scott Huffman

Abstract: Determining a mosquito population’s species composition and age is crucial for estimating therisk of pathogen transmission. At present, age-grading methods are chiefly physiologic andclassify the mosquitoes in terms of parity (e.g., nulliparous or parous). Less commonly usedchronologic methods (e.g., qPCR or near infrared spectroscopy [NIR]) have limited temporalresolution (NIR) or require consumable reagents and technological expertise with molecularmethods. The current lack of robust methods to rapidly evaluate a population’s chronologic agelimits our ability to assess pathogen transmission risk in the context of vectorial capacity estimations (i.e., daily survivability). Our current research seeks to develop methods ofmosquito age determination utilizing mid-infrared spectroscopy and advanced numerical analysis(chemometrics). Infrared (IR) spectroscopy is a type of vibrational spectroscopy that is bothsensitive and information rich. Subtle changes in IR spectra correlate with changes in thebiochemistry of mosquitoes as they age. It has been shown that mosquito species can beidentified using mid infrared spectroscopy and chemometrics. Using mid-infrared spectroscopyand chemometrics, the chronologic age of Aedes triseriatus mosquitoes were predicted usingPLSR and ANN models. Aedes triseriatus were successfully reared into groups of different ageswith low uncertainty in the age. Aedes triseriatus spectra were used to create a training dataset and fit models for prediction using PLSR and ANN. PLSR and ANN models were used to predict the age of samples using a validation dataset with SEPsv of 4.3 and 3.3 days respectively. Mean spectra for each age group were used to try and discern a specific chemical underpinning for the performance of these models and to explain why mosquito age could be predicted using PLSR and ANN models. Peaks between 1200 – 1000 cm-1 typically associated with chitin were investigated and the second derivative of mean absorbance by age at 1032 cm-1increased linearly with age.

Additional Information

Publication
Thesis
Language: English
Date: 2020
Keywords
Chemometrics, Infrared, Mosquito, Spectroscopy

Email this document to