Study Of Molecular Identification In Water Deer (Hydropotes Inermis) And Deer Species

Deer are very important in biodiversity and ecosystem.As a result of the destruction of the ecological environment,habitat loss and illegal hunting and other factors,the lots of wild populations are declining in the rapid,and some have even extinct,many deer species have been classified as endangered species.The protection of water deer and the other deer animal has not attracted enough attention.DNA barcode can identified a query set from an unknown sample to known species,but it is needs reliable arithmetic.Because of different organisms,DNA locus and sequence length,DNA barcoding have different ability and accuracy in species identification.We take the water deer protection for the purpose of molecular recognition of deer species.Provide an effective help for identification of the products from deer species,such as fur,meat and pilos antler.This study uses the specimen of feces,hair and other non-invasive sampling,extraction of genomic DNA,the mitochondrial cytochrome c oxidase subunit I(COI)sequence as a DNA barcode and cytochrome b gene marker recognition etc.We analyze phylogenetic status of water deer.Thirty samples from the 8 deer species were sequenced for COI and Cytb genes,the length of sequences were 700 bp and 792 bp,respectively.Forty three COI sequences of 13 deer species were downloaded from BOLD and GenBank as reference library.According to the related theory and algorithm 30 sequenced samples were assigned to species.A total of 11 kinds of application analysis methods contain automatic barcode gap discovery,neighbor joining,Barcoding with LOGic formulas,nearest neighbour,decision tree,rule-based RIPPER,random forest,support vector machine,artificial intelligence-based,fuzzy set-based and Bayesian-based.The results indicated that there are different ability and accuracy for deer species identification.Both of random forest and support vector machine method are the most reliable for DNA barcoding of deer species classification.The most important factor influencing the performance of the various methods was polymorphism of DNA sequences.The expected performance of most methods can be improved by increasing locus of genes or the length of sequence and the sampling size.In addition,the best-fit analysis method for species classification is selected to improve analysis ability.The coalescent sequences of COI and Cytb were used to constructed phylogenetic tree with Maxmium likelihood and Bayesian inference,in order to investigate the influence on performance of species identification by genes locus and length of the sequence,and to assess system status of Hydropotes inermis.There are similar topology tree with the tree of constructed only COI sequence,but with higher value of bootstrap.Water deer belong to relatively ancient in deer species and it is a closer genetic relationship with roe deer.The results of this study,both of populations of water deer from southeast China and Korean peninsula were not separated,that is,cannot were divided into Hydropotes inermis inermis and Hydropotes inermis argyropus.This may reflect the limitations of DNA barcoding analysis at the subspecies level,or the two geographical subspecies diverge later.