An Efficient Genome Sequence Function Prediction using LDPC Decoding Algorithm

Abstract

Current scenario mainly focuses disease based on genetic. Hence Genome sequence research field concentrated to find a solution for gene sequence analysis and gene function prediction. Now Signal processing field invade this domain and make ease to get optimized solution for the Genetic problem. The proposed work is predicted the Gene function with the help of information coding theory and Neural network concepts. Extended min sum product algorithm (EMS) is used to decode the LDPC (Low density parity check code) and is proven its BER performance near to Shannon limit. On other hand, Extreme learning machine (ELM) and SOM (Self organizing map) in neural network plays vital role in pattern recognition with high speed and less number neurons. So the integration of the two algorithms going to revolute in Gene function prediction and sequence detection. The proposed work is exploited the EMS algorithm efficiency to decode the Genetic ATPG information. DNA Group testing treated analogously to the stochastic Hopfield network with SOM which will reduce error in DNA library. Moreover the work is projected on the lower bounds of DNA word set, results improvement of DNA sequence detection from pooling experiment.