Molecular diet analysis of the house-farm swiftlets (apodidae, collocaliini) in Perak, Malaysia

Chan, Kok Sim (2019) Molecular diet analysis of the house-farm swiftlets (apodidae, collocaliini) in Perak, Malaysia. Master dissertation/thesis, UTAR.

Preview

PDF Download (3417Kb) | Preview

Abstract

House-farming of the edible white-nest swiftlets has been a lucrative industry in Malaysia since three decades ago. This form of semi-captive farming allows the swiftlets to forage outside the swiftlet houses. While it is generally known that swiftlets are insectivores that feed opportunistically, studies on the feeding biology and the diet profiles of these birds are scarce. The present study aims to assess the diet profiles of the house-farm swiftlets using high-throughput sequencing approach, followed by comparison of the diet profiles in different landscapes in Perak, Malaysia. A preliminary assessment of two sets of metabarcoding mitochondrial COI (Cytochrome-c oxidase subunit I) primers was conducted using DNA cloning on the freshly collected swiftlet faeces. The screening indicated that the mICoIintF/HCO2198 primer-pair (for “mICo” region) showed bias towards feather mites, fungal contaminants and swiftlet DNA. Conversely, the LepF1/MLepF1_Rev primer-pair (for “Lep” region) was able to amplify the partial COI sequence of various arthropods in the swiftlet faeces. A bioinformatic pipeline was developed based on the cloning results and can be summarised into four steps for taxonomic assignment of arthropod COI region. High-throughput sequencing using LepF1/MLepF1_Rev for the 218-bp mitochondrial COI region was then performed for the swiftlet faecal samples collected from six swiftlet farms that represented three landscape types, i.e., monocrop, urban and mixed-used landscape. A total of 4,852 operational taxonomic units (OTUs) were generated, out of which 266 were arthropod DNA sequences. Following the taxonomic assignment pipeline established in the preliminary assessment, the diet of the house-farm swiftlets comprised of Diptera (62.74%), followed by Hemiptera (18.87%), Coleoptera (12.26%), Lepidoptera (2.36%), Hymenoptera (1.89%), Blattodea (0.94%) and Odonata (0.94%). A total of 20.30% of OTUs showed no genetic affinity with their respective top BLASTn hits and were therefore of uncertain identity. This could possibly be due to the short length of the DNA marker used in distinguishing the arthropod identity. Furthermore, due to the incompleteness of the DNA database for the insects found in Malaysia, the insect taxa identified using molecular method should also be cross-checked with the species distribution record. The urban (Ipoh) and mixed-use (Pantai Remis) landscape had the most diverse arthropod orders (total of five), while the monocrop landscapes (Beruas OP1 and Beruas OP2) were shown to have the least arthropod orders (total of two). It was proposed that some of the habitats (man-made or natural) in the urban landscape of Ipoh could act as an urban green space to support the substantial arthropod diversity which could be fed upon by the house-farm swiftlets. On the other hand, the mosaic and heterogeneous mixed-use landscape could provide continuity of food source to the house-farm swiftlets. In the monocrop landscapes, the low arthropod abundance in the diet of house-farm swiftlets could be due to some of the farming practices such as the application of pesticides. Also, this study suggests that the urban and mixed-used landscape could be a relatively more ideal place for the house-farm swiftlets to persist. The role of house-farm swiftlets to provide ecological service as insect pest predators in all three landscape types should be further investigated.

Actions (login required)