Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Study on comparison of biochemistry between Trogoderma granarium Everts and Trogoderma variabile Ballion

Al-Shuwaili, Thamer (2020) Study on comparison of biochemistry between Trogoderma granarium Everts and Trogoderma variabile Ballion. PhD thesis, Murdoch University.

PDF - Whole Thesis
Download (3MB) | Preview


Stored grains are paramount commodities to be preserved and stocked for future supply to the market according to the requirement. However, one of the major problems during storage is insect pests, of which insects from Trogoderma sp. especially khapra beetle (Trogoderma granarium) is considered the world most dangerous stored grain insect pests. Therefore, it has been listed as quarantine insect pests in many counties. For timely management of quarantine pest, effective and rapid diagnostic methods are required. Until now, diagnostic technology is mainly based on morphology of insects which require trained taxonomists. Recently, diagnostics based on metabolites and hyperspectral imaging coupled with machine learning is gaining importance. However, very little is known about the metabolites in Trogoderma sp. and how the host grain, gender, and geographical distribution affect the metabolomic profiling in these species is still unknown.

In this thesis, volatile organic compounds (VOCs) emitted by Trogoderma variabile at different life stages were analysed as biomarkers which can help us to understand the biochemistry and metabolomic. Some compounds were identified from T. variabile different stages, which could be used as diagnostic tool for this insect. Gas chromatography coupled to mass spectrometry (GC–MS) was used as a technique to study the metabolite profile of T. variabile in different host grains. However, there are several factors that affect the volatile organic compounds including extraction time and number of insects. The results indicated that the optimal number of insects required for volatile organic compounds (VOC) extraction at each life stage was 25 and 20 for larvae and adults respectively. Sixteen hours were selected as the optimal extraction time for larvae and adults. Some of the VOCs compounds identified from this insect can be used as biomarkers such as pentanoic acid; diethoxymethyl acetate; 1-decyne; naphthalene, 2-methyl-; n-decanoic acid; dodecane, 1-iodo- and m-camphorene from larvae. While butanoic acid, 2-methyl-; pentanoic acid; heptane, 1,1'-oxybis- 2(3H)-Furanone, 5-ethyldihydro-; pentadecane, 2,6,10-trimethyl-; and 1,14-tetradecanediol VOCs, were found in male, whereas pentadecane; nonanic acid; pentadecane, 2,6,10-trimethyl-; undecanal and hexadecanal were identified from female.

Additionaly, direct immersion-solid phase microextraction (DI-SPME) was employed, followed by gas chromatography mass spectrometry analysis (GC-MS) for the collection, separation, and identification of the chemical compounds from T. variabile adults fed on four different host grains. Results showed that insect host grains have a significant difference on the chemical compounds that were identified from female and male. There were 23 compounds identified from adults reared on canola and wheat. However, there were 26 and 28 compounds detected from adults reared on oats and barley respectively. Results showed that 11-methylpentacosane; 13-methylheptacosane; heptacosane; docosane, 1-iodo- and nonacosane were the most significant compounds that identified form T. variabile male reared on different host grains. However, the main compounds identified from female cultured on different host grains include docosane, 1-iodo-; 1-butanamine, N-butyl-; oleic acid; heptacosane; 13-methylheptacosane; hexacosane; nonacosane; 2-methyloctacosane; n-hexadecanoic acid and docosane.

A novel diagnostic tool to identify between T. granarium and T. variabile were developed using visible near infrared hyperspectral imaging and deep learning models including Convolutional Neural Networks (CNN) and Capsule Network. Ventral orientation showed a better accuracy over dorsal orientation of the insects for both larvae and adult stages. This technology offers a new approach and possibility of an effective identification of T. granarium and T. variabile. from its body fragments and larvae skins. The results showed high accuracy to identify between T. granarium and T. variabile. The accuracy was 93.4 and 96.2% for adults and larvae respectively, and the accuracies of 91.6, 91.7 and 90.3% were achieved for larvae skin, adult fragments, larvae fragment respectively.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): Agricultural Sciences
Supervisor(s): Agarwal, Manjree and Ren, Yonglin
Item Control Page Item Control Page


Downloads per month over past year