Research Highlight

AI to identify mosquitoes

Biplab Das

doi:10.1038/nindia.2008.334 Published online 10 December 2008

Researchers have resorted to artificial neural network (ANN) to identify mosquito species that cause diseases like malaria, filaria, Japanese encephalitis and dengue1. Using ANN that mimics the human brain's ability to learn and conclude from experience, the researchers homed in on a specific gene sequence of the mosquito species revealing the mosquito's identity. This tool will be very handy in developing countries where mosquito-borne diseases wreak havoc.

The researchers chose Anopheles mosquito species for their study. They used two different ANN — multi-input single-output neural network (MISONN) and multi-input multi-output neural network (MIMONN) to identify mosquito species by tracking a ribosomal DNA sequence called internal transcribed spacer 2 (ITS2) region of 18 species of Anopheles mosquito.

ITS2 ribosomal DNA sequences of mosquito species are made up of four bases — A (adenine), C (cytosine), T (thymine), and G (guanine). The two types of ANN were trained with inputs of this DNA sequence in binary digits and the output values gave real numbers for the detection of specific mosquito species.

The method of MISONN was found to provide accurate classification of mosquito species based on ITS2 ribosomal DNA sequences. The method presented in this study can be used for any kind of biological classification in genetic level.


  1. Banerjee, K. A. et al. Classification and identification of mosquito species using artificial neural networks. Comput. Biol. Chem. 32, 442-447 (2008) | Article |