Search

Discovering DNA-protein binding sites, also known as motif discovery, is the foundation for further analysis of transcription factors (TFs). Deep learning algorithms such as convolutional neural networks (CNN) have been introduced to motif discovery task and have achieved state-of-art performance. However, due to the limitations of CNN, motif discovery methods based on CNN do not take full advantage of large-scale sequencing data generated by high-throughput sequencing technology. Hence, in this paper we propose multi-scale capsule network architecture (MSC) integrating multi-scale CNN, a variant of CNN able to extract motif features of different lengths, and capsule network, a novel type of artificial neural network architecture aimed at improving CNN. The proposed method is tested on real ChIP-seq datasets and the experimental results show a considerable improvement compared with two well-tested deep learning-based sequence model, DeepBind and Deepsea.

Transcription factors (TFs) play an important role in regulating gene expression, thus the identification of the sites bound by them has become a fundamental step for molecular and cellular biology. In this paper, we developed a deep learning framework leveraging existing fully convolutional neural networks (FCN) to predict TF-DNA binding signals at the base-resolution level, called FCNsignal. The proposed FCNsignal can simultaneously achieve the following tasks: (i) modeling the base-resolution signals of binding regions; (ii) discriminating binding or non-binding regions; (iii) locating TF-DNA binding regions; (iv) predicting binding motifs. The experimental results on 53 TF ChIP-seq datasets and 6 chromatin accessibility ATAC-seq datasets show that our proposed framework outperforms some existing state-of-the-art methods. In addition, we explored to use the trained FCNsignal to locate all potential TF-DNA binding regions on a whole chromosome and predict DNA sequences of arbitrary length, and the results show that our framework can find most of the known binding regions and accept sequences of arbitrary length. Furthermore, we demonstrated the potential ability of our framework in discovering causal disease-associated single-nucleotide polymorphisms (SNPs) through a series of experiments.Author summaryIdentification of transcription factor binding sites (TFBSs) is fundamental to study gene regulatory networks in biological systems, as TFs activate or suppress the transcription of genes by binding to specific TFBSs. With the development of high-throughput sequencing technologies and deep learning (DL), several DL-based approaches have been developed for systematically studying TFBSs, achieving impressive performance. Nevertheless, these methods either excessively focus on discriminating binding or non-binding sequences or individually accomplish multiple TFBSs-associated tasks. In this work, we provide an integrated framework, which utilizes the FCN architecture to predict TF-DNA binding signals at the base-resolution level, to simultaneously study multiple TFBSs-associated tasks. More importantly, we also demonstrate that our proposed framework has the ability to locate all potential TF-DNA binding regions from DNA sequences of arbitrary length. We hope that our framework can provide a new perspective on studying the mechanism of TF-DNA binding and its related tasks.

Deciphering the relationship between transcription factors (TFs) and DNA sequences is very helpful for computational inference of gene regulation and a comprehensive understanding of gene regulation mechanisms. Transcription factor binding sites (TFBSs) are specific DNA short sequences that play a pivotal role in controlling gene expression through interaction with TF proteins. Although recently many computational and deep learning methods have been proposed to predict TFBSs aiming to predict sequence specificity of TF-DNA binding, there is still a lack of effective methods to directly locate TFBSs. In order to address this problem, we propose FCNGRU combing a fully convolutional neural network (FCN) with the gated recurrent unit (GRU) to directly locate TFBSs in this paper. Furthermore, we present a two-task framework (FCNGRU-double): one is a classification task at nucleotide level which predicts the probability of each nucleotide and locates TFBSs, and the other is a regression task at sequence level which predicts the intensity of each sequence. A series of experiments are conducted on 45 in-vitro datasets collected from the UniPROBE database derived from universal protein binding microarrays (uPBMs). Compared with competing methods, FCNGRU-double achieves much better results on these datasets. Moreover, FCNGRU-double has an advantage over a single-task framework, FCNGRU-single, which only contains the branch of locating TFBSs. In additionwe combine with in vivo datasets to make a further analysis and discussion. The source codes are avaiable at https://github.com/wangguoguoa/FCNGRU.

Although Deep learning algorithms have outperformed conventional methods in predicting the sequence specificities of DNA-protein binding, they lack to consider the dependencies among nucleotides and the diverse binding lengths for different transcription factors (TFs). To address the above two limitations simultaneously, in this paper, we propose a high-order convolutional neural network architecture (HOCNN), which employs a high-order encoding method to build high-order dependencies among nucleotides, and a multi-scale convolutional layer to capture the motif features of different length. The experimental results on real ChIP-seq datasets show that the proposed method outperforms the state-of-the-art deep learning method (DeepBind) in the motif discovery task. In addition, we provide further insights about the importance of introducing additional convolutional kernels and the degeneration problem of importing high-order in the motif discovery task.

Transcription factors (TFs) play an important role in regulating gene expression, thus identification of the regions bound by them has become a fundamental step for molecular and cellular biology. In recent years, an increasing number of deep learning (DL) based methods have been proposed for predicting TF binding sites (TFBSs) and achieved impressive prediction performance. However, these methods mainly focus on predicting the sequence specificity of TF-DNA binding, which is equivalent to a sequence-level binary classification task, and fail to identify motifs and TFBSs accurately. In this paper, we developed a fully convolutional network coupled with global average pooling (FCNA), which by contrast is equivalent to a nucleotide-level binary classification task, to roughly locate TFBSs and accurately identify motifs. Experimental results on human ChIP-seq datasets show that FCNA outperforms other competing methods significantly. Besides, we find that the regions located by FCNA can be used by motif discovery tools to further refine the prediction performance. Furthermore, we observe that FCNA can accurately identify TF-DNA binding motifs across different cell lines and infer indirect TF-DNA bindings.

DNA/RNA motif mining is the foundation of gene function research. The DNA/RNA motif mining plays an extremely important role in identifying the DNA- or RNA-protein binding site, which helps to understand the mechanism of gene regulation and management. For the past few decades, researchers have been working on designing new efficient and accurate algorithms for mining motif. These algorithms can be roughly divided into two categories: the enumeration approach and the probabilistic method. In recent years, machine learning methods had made great progress, especially the algorithm represented by deep learning had achieved good performance. Existing deep learning methods in motif mining can be roughly divided into three types of models: convolutional neural network (CNN) based models, recurrent neural network (RNN) based models, and hybrid CNN–RNN based models. We introduce the application of deep learning in the field of motif mining in terms of data preprocessing, features of existing deep learning architectures and comparing the differences between the basic deep learning models. Through the analysis and comparison of existing deep learning methods, we found that the more complex models tend to perform better than simple ones when data are sufficient, and the current methods are relatively simple compared with other fields such as computer vision, language processing (NLP), computer games, etc. Therefore, it is necessary to conduct a summary in motif mining by deep learning, which can help researchers understand this field.

Identifying transcription factor (TF) binding sites (TFBSs) has play an important role in the computational inference of gene regulation. With the development of high-throughput technologies, there have been many conventional methods and deep learning models used in the identification of TFBSs. However, most methods are designed to predict TFBSs only based on raw DNA sequence leads to low accuracy. Therefore, we propose a Dual-channel Convolutional neural network (CNN) model combining DNA sequences and DNA Shape features to predict TFBSs, named DCDS. In the DCDS model, the convolution layer captures low-level features from input data and parallel pooling operations are used to find the most significant activation signal in a sequence for each filter to improve the prediction accuracy of TFBSs. We conduct a series of experiments on 66 in vitro datasets and experimental results show that proposed model DCDS is superior to some state-of-the-art methods.

Classification of functional genomic regions (such as promoters or enhancers) based on sequence data alone is a very important problem. Various data mining algorithms can be used well to apply to predict the promoter region. For example, association and clustering algorithms like Classification And Regression Tree (CART), machine learning algorithms like Simple Logistic, BayesNet, Random forest, or the most popular deep learning like Recurrent Neural Network (RNN), Convolutional Neural Networks (CNN). However, due to large amount of genetic data are unlabeled, these methods cannot directly solve this challenge. Therefore, we present a three-layer dynamic transfer learning language model (TLDTLL) for E. coli promoter classification problems. TLDTLL is an effective algorithm for inductive transfer learning that utilizes pre-training on large unlabeled genomic corpuses. This is particularly advantageous in the context of genomics data, which tends to contain significant volumes of unlabeled data. TLDTLL shows improved results over existing methods for classification of E. coli promoters using only sequence data.

With the rapid development of deep learning, convolution neural network achieve great success in predicting DNA-transcription factor binding, aka motif discovery, In this paper, we propose a novel neural network based architecture i.e. eDeepCNN, combining multi-layer convolution network and embedding layer for predicting in-vitro DNA protein binding sequence. Our model fully utilize fitting capacity of deep convolution neural network and is well designed to capture the interaction pattern between motifs in neighboring sequence. Meanwhile continuous embedding vector serves as a better description of nucleotides than one-hot image-like representation owing to its superior expressive ability. We verify the effectiveness of our model on 20 motif datasets from in-vitro protein binding microarray data (PBMs) and present promising results compared with well-established DeepBind model. In addition, we emphasis the significance of dropout strategy in our model to fight against the overfitting problem along with growing model complexity.

With the rapid development of deep learning, some discriminative motif discovery methods based on deep neural network are gradually becoming the mainstream, which also bringing huge improvement of prediction accuracy. In this paper, we propose a convolutional neural network based architecture (eCNN), combining embedding layer with GloVe. Firstly, eCNN divides each single sequence of ChIP-seq datasets into multiple subsequences called k-mers by a sliding window, and then encoding k-mers into a relatively low dimension vectors by GloVe, and finally scores each vector using multiple convolutional networks. The experiment shows that our architecture can get good results on the task of motif discovery.

Although convolutional neural networks (CNN) have outperformed conventional methods in predicting the sequence specificities of protein-DNA binding in recent years, they do not take full advantage of the intrinsic weakly-supervised information of DNA sequences that a bound sequence may contain multiple TFBS(s). Here, we propose a weakly-supervised convolutional neural network architecture (WSCNN), combining multiple-instance learning (MIL) with CNN, to further boost the performance of predicting protein-DNA binding. WSCNN first divides each DNA sequence into multiple overlapping subsequences (instances) with a sliding window, and then separately models each instance using CNN, and finally fuses the predicted scores of all instances in the same bag using four fusion methods, including Max , Average , Linear Regression , and Top-Bottom Instances . The experimental results on in vivo and in vitro datasets illustrate the performance of the proposed approach. Moreover, models built on in vitro data using WSCNN can predict in vivo protein-DNA binding with good accuracy. In addition, we give a quantitative analysis of the importance of the reverse-complement mode in predicting in vivo protein-DNA binding, and explain why not directly use advanced pooling layers to combine MIL with CNN, through a series of experiments.