“…However, recent advancements in machine learning and deep learning have successfully leveraged genomic data. To date, many groups have successfully constructed promoter prediction tools using traditional machine learning methods, knowledge-based position matrix weight ( Huerta and Collado-Vides, 2003 ; Burden et al, 2005 ; Rangannan and Bansal, 2010 ; Di Salvo et al, 2018 ) through support vector machines, and artificial neural networks for this logistic regression task ( Gordon et al, 2003 ; da Silva et al, 2006 ; Mann et al, 2007 ; Towsey et al, 2008 ; He et al, 2018 ; Liu et al, 2018 ; Rahman et al, 2019 ; Xiao et al, 2019 ; Zhang et al, 2019 ; Li et al, 2021 ). Convolutional neural networks (CNN) and recurrent neural network (RNN)-based architectures (long short-term memory, gated recurrent units) have recently become the most popular choices for promoter classification ( Nguyen et al, 2016 ; Le et al, 2019 ; Oubounyt et al, 2019 ; Amin et al, 2020 ; Zhu et al, 2021 ).…”