An Integrative Framework for Combining Sequence and Epigenomic Data to Predict Transcription Factor Binding Sites Using Deep Learning

Predicting the Dynamic Interaction of Intrinsically Disordered Proteins

Intrinsically disordered proteins (IDPs) participate in various biological processes. Interactions involving IDPs are usually dynamic and are affected by their inherent conformation fluctuations. Comprehensive characterization of these interactions based on current techniques is challenging. Here, we present GSALIDP, a GraphSAGE-embedded LSTM network, to capture the dynamic nature of IDP-involved interactions and predict their behaviors. This framework models multiple conformations of IDP as a dynamic graph, which can effectively describe the fluctuation of its flexible conformation. The dynamic interaction between IDPs is studied, and the data sets of IDP conformations and their interactions are obtained through atomistic molecular dynamic (MD) simulations. Residues of IDP are encoded through a series of features including their frustration. GSALIDP can effectively predict the interaction sites of IDP and the contact residue pairs between IDPs. Its performance in predicting IDP interactions is on par with or even better than the conventional models in predicting the interaction of structural proteins. To the best of our knowledge, this is the first model to extend the protein interaction prediction to IDP-involved interactions.

Prediction of Transcription Factor Binding Sites With an Attention Augmented Convolutional Neural Network

Identification of transcription factor binding sites (TFBSs) is essential for revealing the rules of protein-DNA binding. Although some computational methods have been presented to predict TFBSs using epigenomic and sequence features, most of them ignore the common features among cross-cell types. It is still unclear to what extent the common features could help for this task. To this end, we proposed a new method (named Attention-augmented Convolutional Neural Network, or ACNN) to predict TFBSs. ACNN uses attention-augmented convolutional layers to capture global and local contexts in DNA sequences and employs the convolutional layers to capture features of histone modification markers. In addition, ACNN adopts the private and shared convolutional neural network (CNN) modules to learn specific and common features, respectively. To encourage the shared CNN module to learn the common features, adversarial training is applied in ACNN. The results on 253 ChIP-seq datasets show that ACNN outperforms other existing methods. The attention-augmented convolutional layers and adversarial training mechanism in ACNN can effectively improve the prediction performance. Moreover, in the case of limited labeled data, ACNN also performs better than a baseline method. We further visualize the convolution kernels as motifs to explain the interpretability of ACNN.

Genomics enters the deep learning era

The tremendous amount of biological sequence data available, combined with the recent methodological breakthrough in deep learning in domains such as computer vision or natural language processing, is leading today to the transformation of bioinformatics through the emergence of deep genomics, the application of deep learning to genomic sequences. We review here the new applications that the use of deep learning enables in the field, focusing on three aspects: the functional annotation of genomes, the sequence determinants of the genome functions and the possibility to write synthetic genomic sequences.

InsuLock: A Weakly Supervised Learning Approach for Accurate Insulator Prediction, and Variant Impact Quantification

Mapping chromatin insulator loops is crucial to investigating genome evolution, elucidating critical biological functions, and ultimately quantifying variant impact in diseases. However, chromatin conformation profiling assays are usually expensive, time-consuming, and may report fuzzy insulator annotations with low resolution. Therefore, we propose a weakly supervised deep learning method, InsuLock, to address these challenges. Specifically, InsuLock first utilizes a Siamese neural network to predict the existence of insulators within a given region (up to 2000 bp). Then, it uses an object detection module for precise insulator boundary localization via gradient-weighted class activation mapping (~40 bp resolution). Finally, it quantifies variant impacts by comparing the insulator score differences between the wild-type and mutant alleles. We applied InsuLock on various bulk and single-cell datasets for performance testing and benchmarking. We showed that it outperformed existing methods with an AUROC of ~0.96 and condensed insulator annotations to ~2.5% of their original size while still demonstrating higher conservation scores and better motif enrichments. Finally, we utilized InsuLock to make cell-type-specific variant impacts from brain scATAC-seq data and identified a schizophrenia GWAS variant disrupting an insulator loop proximal to a known risk gene, indicating a possible new mechanism of action for the disease.

High-resolution transcription factor binding sites prediction improved performance and interpretability by deep learning method

Transcription factors (TFs) are essential proteins in regulating the spatiotemporal expression of genes. It is crucial to infer the potential transcription factor binding sites (TFBSs) with high resolution to promote biology and realize precision medicine. Recently, deep learning-based models have shown exemplary performance in the prediction of TFBSs at the base-pair level. However, the previous models fail to integrate nucleotide position information and semantic information without noisy responses. Thus, there is still room for improvement. Moreover, both the inner mechanism and prediction results of these models are challenging to interpret. To this end, the Deep Attentive Encoder-Decoder Neural Network (D-AEDNet) is developed to identify the location of TFs-DNA binding sites in DNA sequences. In particular, our model adopts Skip Architecture to leverage the nucleotide position information in the encoder and removes noisy responses in the information fusion process by Attention Gate. Simultaneously, the Transcription Factor Motif Discovery based on Sliding Window (TF-MoDSW), an approach to discover TFs-DNA binding motifs by utilizing the output of neural networks, is proposed to understand the biological meaning of the predicted result. On ChIP-exo datasets, experimental results show that D-AEDNet has better performance than competing methods. Besides, we authenticate that Attention Gate can improve the interpretability of our model by ways of visualization analysis. Furthermore, we confirm that ability of D-AEDNet to learn TFs-DNA binding motifs outperform the state-of-the-art methods and availability of TF-MoDSW to discover biological sequence motifs in TFs-DNA interaction by conducting experiment on ChIP-seq datasets.

Regulatory SNPs: Altered Transcription Factor Binding Sites Implicated in Complex Traits and Diseases

The vast majority of the genetic variants (mainly SNPs) associated with various human traits and diseases map to a noncoding part of the genome and are enriched in its regulatory compartment, suggesting that many causal variants may affect gene expression. The leading mechanism of action of these SNPs consists in the alterations in the transcription factor binding via creation or disruption of transcription factor binding sites (TFBSs) or some change in the affinity of these regulatory proteins to their cognate sites. In this review, we first focus on the history of the discovery of regulatory SNPs (rSNPs) and systematized description of the existing methodical approaches to their study. Then, we brief the recent comprehensive examples of rSNPs studied from the discovery of the changes in the TFBS sequence as a result of a nucleotide substitution to identification of its effect on the target gene expression and, eventually, to phenotype. We also describe state-of-the-art genome-wide approaches to identification of regulatory variants, including both making molecular sense of genome-wide association studies (GWAS) and the alternative approaches the primary goal of which is to determine the functionality of genetic variants. Among these approaches, special attention is paid to expression quantitative trait loci (eQTLs) analysis and the search for allele-specific events in RNA-seq (ASE events) as well as in ChIP-seq, DNase-seq, and ATAC-seq (ASB events) data.

Computational methods for the prediction of chromatin interaction and organization using sequence and epigenomic profiles

The exploration of three-dimensional chromatin interaction and organization provides insight into mechanisms underlying gene regulation, cell differentiation and disease development. Advances in chromosome conformation capture technologies, such as high-throughput chromosome conformation capture (Hi-C) and chromatin interaction analysis by paired-end tag (ChIA-PET), have enabled the exploration of chromatin interaction and organization. However, high-resolution Hi-C and ChIA-PET data are only available for a limited number of cell lines, and their acquisition is costly, time consuming, laborious and affected by theoretical limitations. Increasing evidence shows that DNA sequence and epigenomic features are informative predictors of regulatory interaction and chromatin architecture. Based on these features, numerous computational methods have been developed for the prediction of chromatin interaction and organization, whereas they are not extensively applied in biomedical study. A systematical study to summarize and evaluate such methods is still needed to facilitate their application. Here, we summarize 48 computational methods for the prediction of chromatin interaction and organization using sequence and epigenomic profiles, categorize them and compare their performance. Besides, we provide a comprehensive guideline for the selection of suitable methods to predict chromatin interaction and organization based on available data and biological question of interest.

A machine learning framework for the prediction of chromatin folding in Drosophila using epigenetic features

Technological advances have lead to the creation of large epigenetic datasets, including information about DNA binding proteins and DNA spatial structure. Hi-C experiments have revealed that chromosomes are subdivided into sets of self-interacting domains called Topologically Associating Domains (TADs). TADs are involved in the regulation of gene expression activity, but the mechanisms of their formation are not yet fully understood. Here, we focus on machine learning methods to characterize DNA folding patterns in Drosophila based on chromatin marks across three cell lines. We present linear regression models with four types of regularization, gradient boosting, and recurrent neural networks (RNN) as tools to study chromatin folding characteristics associated with TADs given epigenetic chromatin immunoprecipitation data. The bidirectional long short-term memory RNN architecture produced the best prediction scores and identified biologically relevant features. Distribution of protein Chriz (Chromator) and histone modification H3K4me3 were selected as the most informative features for the prediction of TADs characteristics. This approach may be adapted to any similar biological dataset of chromatin features across various cell lines and species. The code for the implemented pipeline, Hi-ChiP-ML, is publicly available: https://github.com/MichalRozenwald/Hi-ChIP-ML.

Prediction of enhancer-promoter interactions using the cross-cell type information and domain adversarial neural network

BACKGROUND

Enhancer-promoter interactions (EPIs) play key roles in transcriptional regulation and disease progression. Although several computational methods have been developed to predict such interactions, their performances are not satisfactory when training and testing data from different cell lines. Currently, it is still unclear what extent a across cell line prediction can be made based on sequence-level information.

RESULTS

In this work, we present a novel Sequence-based method (called SEPT) to predict the enhancer-promoter interactions in new cell line by using the cross-cell information and Transfer learning. SEPT first learns the features of enhancer and promoter from DNA sequences with convolutional neural network (CNN), then designing the gradient reversal layer of transfer learning to reduce the cell line specific features meanwhile retaining the features associated with EPIs. When the locations of enhancers and promoters are provided in new cell line, SEPT can successfully recognize EPIs in this new cell line based on labeled data of other cell lines. The experiment results show that SEPT can effectively learn the latent import EPIs-related features between cell lines and achieves the best prediction performance in terms of AUC (the area under the receiver operating curves).

CONCLUSIONS

SEPT is an effective method for predicting the EPIs in new cell line. Domain adversarial architecture of transfer learning used in SEPT can learn the latent EPIs shared features among cell lines from all other existing labeled data. It can be expected that SEPT will be of interest to researchers concerned with biological interaction prediction.

Inferring subgroup-specific driver genes from heterogeneous cancer samples via subspace learning with subgroup indication

MOTIVATION

Detecting driver genes from gene mutation data is a fundamental task for tumorigenesis research. Due to the fact that cancer is a heterogeneous disease with various subgroups, subgroup-specific driver genes are the key factors in the development of precision medicine for heterogeneous cancer. However, the existing driver gene detection methods are not designed to identify subgroup specificities of their detected driver genes, and therefore cannot indicate which group of patients is associated with the detected driver genes, which is difficult to provide specifically clinical guidance for individual patients.

RESULTS

By incorporating the subspace learning framework, we propose a novel bioinformatics method called DriverSub, which can efficiently predict subgroup-specific driver genes in the situation where the subgroup annotations are not available. When evaluated by simulation datasets with known ground truth and compared with existing methods, DriverSub yields the best prediction of driver genes and the inference of their related subgroups. When we apply DriverSub on the mutation data of real heterogeneous cancers, we can observe that the predicted results of DriverSub are highly enriched for experimentally validated known driver genes. Moreover, the subgroups inferred by DriverSub are significantly associated with the annotated molecular subgroups, indicating its capability of predicting subgroup-specific driver genes.

AVAILABILITY AND IMPLEMENTATION

The source code is publicly available at https://github.com/JianingXi/DriverSub.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Local Epigenomic Data are more Informative than Local Genome Sequence Data in Predicting Enhancer-Promoter Interactions Using Neural Networks

Enhancer-promoter interactions (EPIs) are crucial for transcriptional regulation. Mapping such interactions proves useful for understanding disease regulations and discovering risk genes in genome-wide association studies. Some previous studies showed that machine learning methods, as computational alternatives to costly experimental approaches, performed well in predicting EPIs from local sequence and/or local epigenomic data. In particular, deep learning methods were demonstrated to outperform traditional machine learning methods, and using DNA sequence data alone could perform either better than or almost as well as only utilizing epigenomic data. However, most, if not all, of these previous studies were based on randomly splitting enhancer-promoter pairs as training, tuning, and test data, which has recently been pointed out to be problematic; due to multiple and duplicating/overlapping enhancers (and promoters) in enhancer-promoter pairs in EPI data, such random splitting does not lead to independent training, tuning, and test data, thus resulting in model over-fitting and over-estimating predictive performance. Here, after correcting this design issue, we extensively studied the performance of various deep learning models with local sequence and epigenomic data around enhancer-promoter pairs. Our results confirmed much lower performance using either sequence or epigenomic data alone, or both, than reported previously. We also demonstrated that local epigenomic features were more informative than local sequence data. Our results were based on an extensive exploration of many convolutional neural network (CNN) and feed-forward neural network (FNN) structures, and of gradient boosting as a representative of traditional machine learning.