Deep Learning to Identify Transcription Start Sites from CAGE Data

摘要: Gene transcription start site (TSS) identification is important to understanding transcriptional gene regulation. Cap Analysis Expression (CAGE) experiments have recently become common practice for direct measurement of TSSs. Currently, CAGE data available in public databases created unprecedented opportunities study initiation mechanisms under various cellular conditions. However, due potential noises inherent data, in-silico methods are required identify bonafide TSSs from further. Here we present a computational approach dlCAGE, an end-to-end deep neural network data. dlCAGE incorporate de-novo DNA regulatory motif features discovered by DeepBind model architecture, as well existing sequence and structural features. Testing results several cell lines comparison with current state-of-the-art approaches showed its superior performance promise TSS experiments.