Aims The aim of this study was to identify and compare the gut microbial community of wild and captive Tor tambroides through 16S rDNA metagenetic sequencing followed by functions prediction. Methods and results The library of 16S rDNA V3‐V4 hypervariable regions of gut microbiota was amplified and sequenced using Illumina MiSeq. The sequencing data were analyzed using Quantitative Insights into Microbial Ecology ( QIIME ) pipeline and Phylogenetic Investigation of Communities by Reconstruction of Unobserved States ( PICRUS t). The most abundant bacterial phyla in both wild and captive T. tambroides were Firmicutes, Proteobacteria, Fusobacteria and Bacteroidetes. Cetobacterium spp., Peptostreptococcaceae family, Bacteroides spp., Phosphate solubilizing bacteria PSB ‐M‐3, and Vibrio spp. were five most abundant OTU in wild T. tambroides as compared to Cetobacterium spp., Citrobacter spp., Aeromonadaceae family, Peptostreptococcaceae family and Turicibacter spp. in captive T. tambroides . Conclusion In this study, the specimens of the wild T. tambroides contain more diverse gut microbiota than of the captive ones. The results suggested that Cetobacterium spp. is one of the core microbiota in guts of T. tambroides . Besides, high abundant Bacteroides spp., Citrobacter spp., Turicibacter spp., and Bacillus spp. may provide important functions in T. tambroides guts. Significance and impact of the study The results of this study provide significant information of T. tambroides gut microbiota for further understanding of their physiological functions including growth and disease resistance.

Silage produced in tropical countries is prone to spoilage because of high humidity and temperature. Therefore, determining indigenous bacteria as potential inoculants is important to improve silage quality. This study aimed to determine bacterial community and functional changes associated with ensiling using amplicon metagenomics and to predict potential bacterial additives associated with silage quality in the Malaysian climate. Silages of two forage crops (sweet corn and Napier) were prepared, and their fermentation properties and functional bacterial communities were analysed. After ensiling, both silages were predominated by lactic acid bacteria (LAB), and they exhibited good silage quality with significant increment in lactic acid, reductions in pH and water-soluble carbohydrates, low level of acetic acid and the absence of propionic and butyric acid. LAB consortia consisting of homolactic and heterolactic species were proposed to be the potential bacterial additives for sweet corn and Napier silage fermentation. Tax4fun functional prediction revealed metabolic pathways related to fermentation activities (bacterial division, carbohydrate transport and catabolism, and secondary metabolite production) were enriched in ensiled crops (p < 0.05). These results might suggest active transport and metabolism of plant carbohydrates into a usable form to sustain bacterial reproduction during silage fermentation, yielding metabolic products such as lactic acid. This research has provided a comprehensive understanding of bacterial communities before and after ensiling, which can be useful for desirable silage fermentation in Malaysia.

Hand gesture recognition (HGR) is a crucial area of research that enhances communication by overcoming language barriers and facilitating human-computer interaction. Although previous works in HGR have employed deep neural networks, they fail to encode the orientation and position of the hand in the image. To address this issue, this paper proposes HGR-ViT, a Vision Transformer (ViT) model with an attention mechanism for hand gesture recognition. Given a hand gesture image, it is first split into fixed size patches. Positional embedding is added to these embeddings to form learnable vectors that capture the positional information of the hand patches. The resulting sequence of vectors are then served as the input to a standard Transformer encoder to obtain the hand gesture representation. A multilayer perceptron head is added to the output of the encoder to classify the hand gesture to the correct class. The proposed HGR-ViT obtains an accuracy of 99.98%, 99.36% and 99.85% for the American Sign Language (ASL) dataset, ASL with Digits dataset, and National University of Singapore (NUS) hand gesture dataset, respectively.