Operon Finder: A Deep Learning-based Web Server for Accurate Prediction of Prokaryotic Operons

Abstract

Operons are groups of consecutive genes that transcribe together under the regulation of a common promoter. They influence protein regulation and various physiological pathways, making their accurate detection desirable. The detection of operons through experimental means is a laborious and financially intensive process. Therefore, human experts predict potential operons utilizing their prior knowledge of the genetic organization and functional correlation of the genes. However, with the rise in the number of completely sequenced genomes, the development of automated algorithms, tools, and web servers is highly preferred over manual detection for operon prediction. Currently available state-of-the-art algorithms use a deep learning-based model to predict if the adjacent genes belong to the same operons in the given genome. However, these require an understanding of programming knowledge and computational skills, making them not-very user friendly. In this study, we developed a user-friendly web service, Operon Finder, for on-the-fly prediction of operons using the deep learning method. The interface provides a facility for genome search, operon live-filtering, viewing operonic DNA sequences, downloading predicted results, and links for data retrieval from the NCBI (National Center for Biotechnology Information) database. The web server is available at https://www.iitg.ac.in/spkanaujia/operonfinder.html. The experimental methods and the implementation details are publicly available at https://github.com/SPKlab/Operon-Finder.

Keywords: Algorithm; Deep learning; Genome analysis; Operon; Web server.