Design And Implementation Of The Neuropeptide Information Extraction System Based On PubMed Abstract

Currently, neuroscience research is one of the most remarkable discipline in lifescience research, and neuropeptides in neuroscience is one of the fastest progress frontierin recent years, neuropeptide research is leaps and bounds. Neuropeptide exists not only inthe nervous system, the system is also in widespread systemic,therefore it s important tostudy neuropeptide information extraction technology.After a summary of depth analysis of the existing neuropeptide text data,designedand implemented a neuropeptide information extraction system. The system functionmodule consists of three parts: the identification and extraction of the peptide sequence,identifying the name of the neuropeptides, neuropeptide peptide sequences matching thenames and evaluation.First, by studying a large number constitute characteristics ofpeptide sequences in the biological text, designed and implemented a approach to identifyand extract the peptide sequence.Secondly,by extracting from several existing databasewhich have included neuropeptides, and through the analysis of these names wordformation, affixes characteristics, generate a regular expression-based neuropeptidename dictionary for recognition.Then, in the implement of matching evaluation,scoringprocess is based on rules and experience, and the scoring function parameters in the designhave been explained.Finally,design different experiments, experiment results demonstrate that scoreprediction has a good performance and the system performance can meet the actualdemand.