DeepTFactor, a deep learning-based tool for the identification of transcription factors

We report the development of a deep learning-based tool, DeepTFactor, that predicts whether a protein of question is a transcription factor. DeepTFactor uses a convolutional neural network to extract features of protein sequences. We characterized the genome-wide binding sites of three TFs (i.e., YqhC, YiaU, and YahB), which are predicted by DeepTFactor Overall design: Identification of genome-wide bindings for uncharacterized transcription factors in E. coli K-12 MG1655, using ChIP-exo technology

本研究开发了一款基于深度学习的工具DeepTFactor，可用于预测目标蛋白质是否为转录因子（transcription factor）。该工具采用卷积神经网络（convolutional neural network）提取蛋白质序列特征。本研究对3个经DeepTFactor预测得到的转录因子（TFs，即YqhC、YiaU与YahB）的全基因组结合位点进行了表征。实验设计概述：利用ChIP-exo技术，对大肠杆菌（E. coli）K-12 MG1655中未被表征的转录因子的全基因组结合位点进行鉴定。