Classifying Promoters by Interpreting the Hidden Information of DNA Sequences via Deep Learning and Combination of Continuous FastText N-Grams

Abstract

A promoter is a short region of DNA (100-1,000 bp) where transcription of a gene by RNA polymerase begins. It is typically located directly upstream or at the 5' end of the transcription initiation site. DNA promoter has been proven to be the primary cause of many human diseases, especially diabetes, cancer, or Huntington's disease. Therefore, classifying promoters has become an interesting problem and it has attracted the attention of a lot of researchers in the bioinformatics field. There were a variety of studies conducted to resolve this problem, however, their performance results still require further improvement. In this study, we will present an innovative approach by interpreting DNA sequences as a combination of continuous FastText N-grams, which are then fed into a deep neural network in order to classify them. Our approach is able to attain a cross-validation accuracy of 85.41 and 73.1% in the two layers, respectively. Our results outperformed the state-of-the-art methods on the same dataset, especially in the second layer (strength classification). Throughout this study, promoter regions could be identified with high accuracy and it provides analysis for further biological research as well as precision medicine. In addition, this study opens new paths for the natural language processing application in omics data in general and DNA sequences in particular.

Keywords: DNA promoter; convolutional neural network; natural language processing; precision medicine; transcription factor; word embedding.

Figure 1

Process of promoters in transcription. (A) The gene is essentially turned off. The repressor is not inhibited by lactose and binds to operator, then promoter is bound to make lactase; (B) the gene is turned on. The repressor is inhibited by lactose, then the promoter is bound by the RNA polymerase and express the genes to synthesize lactase. Finally, the lactase will digest all of the lactose, until nothing binds to the repressor. The repressor will then bind to the operator, stopping the manufacture of lactase.

Figure 2

Flowchart of this study. First, we used FastText to train model and extract features from benchmark dataset (Xiao et al., 2018), then combined 10-gram levels to a combination sets of vectors (1,000 dimensions). Deep neural network was then constructed to learn these vectors and classify the DNA sequences.

Figure 3

Performance results on identifying promoters using different levels of N-gram. Our classifier could classify promoters with high performance (AUC ~ 0.9), especially at 4-gram and 5-gram levels.