EXON-INTRON pattern analysis
Organizer
Speaker
Time
Monday, October 14, 2024 9:00 PM - 9:30 PM
Venue
Online
Abstract
Provide a new approach to predict the EXON/INTRON position pattern.First, we use NV to training a NN for hs37 dataset. Then we select an existing gene’s EXON and INTRON segment construct the full fasta sequence as the test dataset.The new approach start with a general pattern match according to the rule:每个内含子的开始两个碱基都是GT,最后两个是AG,Then generate all potential possible position patterns.By predicting each segment EXON/INTRON using NN, we can eliminate invalid patterns.Finally further combining with biology info to eliminate more invalid patterns.
Speaker Intro
After graduating with a PhD, he mainly worked in the field of wireless communications. He has worked in Lucent, Alcatel-Lucent, and Nokia as a senior engineer. He has 23 years of rich knowledge and experience in the field of wireless communications. Currently he is a research fellow of Beijing Institute of Mathematical Sciences and Applications (BIMSA) engaged in research on 4G/5G Wireless Communication, Biomathematics, Neural Network, Artificial Intelligence, Big Data, Machine Learning, nonlinear filter.