iPro-TCN: Prediction of DNA Promoters Recognition and Their Strength Using Temporal Convolutional Network

Promoters are important regulatory elements in the genome that control gene expression, and their abnormalities have been linked to various diseases. With development of high-throughput sequencing techniques, there is a need for computational methods identify promoters large amounts DNA sequence data. However, many existing rely on single feature representation approach, which can potentially lead information loss. In order address this issue. We proposed model iPro-TCN combines Temporal Convolutional Network (TCN) with word2vec representation. This includes new descriptor called K-mer word vector has shown high sensitivity accuracy distinguishing promoters, including strong weak promoters. The was evaluated benchmark dataset able achieve good performance both promoter identification strength prediction. Specifically, had an 91.86% first layer identification, 84.63% second These results suggest performer predicting sequences DNA. far higher than best predictor, indicates our showcases better compared approaches.