gkm-DNN: efficient prediction using gapped k-mer features and deep neural networks

Abstract

How to extract informative features from genome sequence is a challenging issue. Gapped k-mers frequency vectors (gkm-fv) has been presented as a new type of features in the last few years. Coupled with support vector machine (gkm-SVM), gkm-fvs have been used to achieve effective sequence-based predictions. However, the huge computation of a large kernel matrix prevents it from using large amount of data. And it is unclear how to combine gkm-fvs with other data sources in the context of string kernel. On the other hand, the high dimensionality, colinearity and sparsity of gkm-fvs hinder the use of many traditional machine learning methods without a kernel trick. Therefore, we proposed a flexible and scalable framework gkm-DNN to achieve feature representation from high-dimensional gkm-fvs using deep neural networks (DNN). We first proposed a more concise version of gkm-fvs which significantly reduce the dimension of gkm-fvs. Then we implemented an efficient method to calculate the gkm-fv of a given sequence at the first time. Finally, we adopted a DNN model with gkm-fvs as inputs to achieve efficient feature representation and a prediction task. Here, we took the transcription factor binding site prediction as an illustrative application. We applied gkm-DNN onto 467 small and 69 big human ENCODE ChIP-seq datasets to demonstrate its performance and compared it with the state-of-the-art method gkm-SVM. We demonstrated that gkm-DNN can not only improve the limitations of high dimensionality, colinearity and sparsity of gkm-fvs, but also make comparable overall performance compared with gkm-SVM using the same gkm-fvs. In addition, we used gkm-DNN to explore the representation power of gkm-fvs and provided more explanation on how gkm-fvs work.