Postgraduate Student from the College of Life Sciences Published Paper in Authoritative International Journal on Evolutionary Biology2021-10-29

Source:       Clicks:       Size: small     middle       big

  Recently, the genetic diversity and evolution team of the College of Life Sciences of CNU proposed a new species identification method based on deep-learning convolutional neural network, constructed a new deep-learning network model (MMNet), and realized the automatic integration of morphological and molecular data for species identification for the first time. The model was tested centrally on real data with extensive representativeness, and the result showed that the proposed MMNet method was significantly better than the common method using just data. In addition, the new MMNet method also featured extremely high accuracy (>98%) in the identification of same and related species, a more challenging test. The new method was robust to sequence length and image size. Further analysis showed that both morphological data and genetic data were important to the model, and genetic data contributed more to the model.

  The study was published online on September 15 in Systematic Biology, a famous international journal on evolutionary biology (IF=15.683, ranking 3/50, Evolutionary Biology; ranking 7/647, Ecology, Evolution, Behavior and Systematics). Yang Bing, postgraduate student from the College of Life Sciences, is the first author of this paper. Associate Professor Zhang Zhenxin, a young teacher from the College of Resource Environment and Tourism, is the co-first and corresponding author of this paper. Professor Zhang Aibing is the final corresponding author of this paper.

  This is the first time for Professor Zhang Aibing to, with CNU as the first unit, guide postgraduate students to publish research results in famous international journals on evolutionary biology since he proposed a new method for species identification by DNA bar code based on artificial intelligence in 2008 (Zhang et al. 2008. Syst. Biol.). The research won the funding of the National Outstanding Youth Fund, the General Project of the National Natural Science Foundation of China, and the Multidisciplinary Program of the Academy for Multidisciplinary Studies of CNU.