Promoter analysis and prediction in the human genome using sequence-based deep learning models
Autor: | Ramzan Umarov, Yu Li, Victor V. Solovyev, Xin Gao, Hiroyuki Kuwahara |
---|---|
Rok vydání: | 2019 |
Předmět: |
Statistics and Probability
Web server Computer science 0206 medical engineering Genomics 02 engineering and technology Computational biology computer.software_genre Biochemistry Genome 03 medical and health sciences Deep Learning Discriminative model Transcription (biology) Humans Promoter Regions Genetic Molecular Biology 030304 developmental biology 0303 health sciences Transcription initiation site Genome Human business.industry Deep learning Promoter Computer Science Applications Computational Mathematics Computational Theory and Mathematics Transcription preinitiation complex Human genome Artificial intelligence Transcription Initiation Site business computer 020602 bioinformatics |
Zdroj: | Bioinformatics. 35:2730-2737 |
ISSN: | 1460-2059 1367-4803 |
Popis: | MotivationComputational identification of promoters is notoriously difficult as human genes often have unique promoter sequences that provide regulation of transcription and interaction with transcription initiation complex. While there are many attempts to develop computational promoter identification methods, we have no reliable tool to analyze long genomic sequences.ResultsIn this work, we further develop our deep learning approach that was relatively successful to discriminate short promoter and non-promoter sequences. Instead of focusing on the classification accuracy, in this work we predict the exact positions of the transcription start site inside the genomic sequences testing every possible location. We studied human promoters to find effective regions for discrimination and built corresponding deep learning models. These models use adaptively constructed negative set, which iteratively improves the model’s discriminative ability. Our method significantly outperforms the previously developed promoter prediction programs by considerably reducing the number of false-positive predictions. We have achieved error-per-1000-bp rate of 0.02 and have 0.31 errors per correct prediction, which is significantly better than the results of other human promoter predictors.Availability and implementationThe developed method is available as a web server at http://www.cbrc.kaust.edu.sa/PromID/. |
Databáze: | OpenAIRE |
Externí odkaz: |