A Review About Transcription Factor Binding Sites Prediction Based on Deep Learning

Zeng, Yuanqi; Gong, Meiqin; Meng, Lin; Gao, Dongrui; Zhang, Yongqing

doi:10.1109/access.2020.3042903

Cited by 24 publications

(10 citation statements)

References 104 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using REPTILE, we pre-processed these datasets to generate two confident sets of active and non-active enhancers. We took advantage of the capabilities of CNNs, which have shown great success in many bioinformatic challenges in recent years ( Barshai et al, 2020 ; Zeng et al, 2020 ; He et al, 2021 ), to learn the important features in sequence data to predict whether a DNA sequence belongs to an active or non-active enhancer region. Moreover, with the trained models and sequence datasets, we used the Integrated Gradient method to highlight the important features in every sample in the dataset and aggregated the results by TF-MoDISco to extract putative regulatory motifs.…”

Section: Discussionmentioning

confidence: 99%

Deciphering transcription factors and their corresponding regulatory elements during inhibitory interneuron differentiation using deep neural networks

Alatawneh

Salomon²,

Eshel³

et al. 2023

Front. Cell Dev. Biol.

View full text Add to dashboard Cite

During neurogenesis, the generation and differentiation of neuronal progenitors into inhibitory gamma-aminobutyric acid-containing interneurons is dependent on the combinatorial activity of transcription factors (TFs) and their corresponding regulatory elements (REs). However, the roles of neuronal TFs and their target REs in inhibitory interneuron progenitors are not fully elucidated. Here, we developed a deep-learning-based framework to identify enriched TF motifs in gene REs (eMotif-RE), such as poised/repressed enhancers and putative silencers. Using epigenetic datasets (e.g., ATAC-seq and H3K27ac/me3 ChIP-seq) from cultured interneuron-like progenitors, we distinguished between active enhancer sequences (open chromatin with H3K27ac) and non-active enhancer sequences (open chromatin without H3K27ac). Using our eMotif-RE framework, we discovered enriched motifs of TFs such as ASCL1, SOX4, and SOX11 in the active enhancer set suggesting a cooperativity function for ASCL1 and SOX4/11 in active enhancers of neuronal progenitors. In addition, we found enriched ZEB1 and CTCF motifs in the non-active set. Using an in vivo enhancer assay, we showed that most of the tested putative REs from the non-active enhancer set have no enhancer activity. Two of the eight REs (25%) showed function as poised enhancers in the neuronal system. Moreover, mutated REs for ZEB1 and CTCF motifs increased their in vivo activity as enhancers indicating a repressive effect of ZEB1 and CTCF on these REs that likely function as repressed enhancers or silencers. Overall, our work integrates a novel framework based on deep learning together with a functional assay that elucidated novel functions of TFs and their corresponding REs. Our approach can be applied to better understand gene regulation not only in inhibitory interneuron differentiation but in other tissue and cell types.

show abstract

Section: Discussionmentioning

confidence: 99%

Deciphering transcription factors and their corresponding regulatory elements during inhibitory interneuron differentiation using deep neural networks

Alatawneh

Salomon²,

Eshel³

et al. 2023

Front. Cell Dev. Biol.

View full text Add to dashboard Cite

show abstract

“…TFs typically contain a DNA-Binding domain (DBD) connected to speci c DNA sequences proximal to regulatory genes [4]. TFBS, with lengths ranging from 4-30 base pairs, are pivotal for precise prediction.…”

Section: Introductionmentioning

confidence: 99%

CBLANE: A deep learning approach for Transcription Factor Binding Sites Prediction

Ferrao,

Dias,

Morajkar

2024

Preprint

View full text Add to dashboard Cite

This study explores the application of CBLANE (convolutional bidirectional long short-term memory (BiLSTM) attention network) as a deep neural network designed for predicting transcription factor binding sites (TFBS) within genomic data. CBLANE's architecture comprises convolutional, recurrent, and attention layers, tailored to extract essential features and information from DNA sequence data. Initially trained on DNA sequences, CBLANE can also function as an encoder, useful for dimensionality reduction and the extraction of information from genetic sequences. Its architecture enables the extraction of relevant features critical for TFBS prediction. Thoroughly evaluating the model, we find that CBLANE has an average AUC of 0.9386 on the 690 datasets from the Encyclopedia of DNA Elements (ENCODE) chromatin immunoprecipitation sequencing (ChIP-seq) experiments outperforming other state of the art methods. Further experiments on the 165 ENCODE ChIP-Seq datasets reveal that CBLANE attains an average AUC of 0.9407. This performance surpasses that of other state-of-the-art methods that incorporate DNA shape profiles in their predictions. Notably, this improved performance was attained while substantially reducing the model size, as reflected in the parameter count.

show abstract

“…The binding sites of PRDM9 can be identified by Chromatin Immuno-Precipitation with high-throughput sequencing (ChIPseq) experiments [8]. However, it is impossible to determine all the binding sites using a limited number of experimental conditions [11], which hampers the comprehensive understanding of 1 Graduate School of Information Science and Technology, Hokkaido University, Sapporo, Hokkaido 060-0814, Japan 2 Faculty of Information Science and Technology, Hokkaido University, Sapporo, Hokkaido 060-0814, Japan a) nosada@ist.hokudai.ac.jp the DNA-protein binding mechanisms. Previous studies showed that the canonical 13-mer binding motif of PRDM9 (CCNCC-NTNNCCNC), which is enriched in hotspots, is insufficient to explain all the hotspots [1], [2], [12], [13].…”

Section: Introductionmentioning

confidence: 99%

Predicting PRDM9 Binding Sites by a Convolutional Neural Network and Verification Using Genetic Recombination Map

Nakamura

Endo

Osada

2022

IPSJ Transactions on Bioinformatics

View full text Add to dashboard Cite

PR domain-containing 9 (PRDM9) is a zinc-finger protein that binds to specific DNA motifs and induces the crossing-over between chromosomes, resulting in a high recombination rate around binding sites. Currently, the binding sites of PRDM9 are predicted with methods based on motif matching and Position-specific Weight Matrix (PWM). Meanwhile, the Convolutional Neural Network (CNN) has shown superior performance in recent studies to identify protein-binding regions in general, and it is expected to perform well in PRDM9 binding site prediction. In this study, we compared the performance of PWM and CNN for predicting PRDM9 binding sites with not only test data but also the correlation between the prediction score for a fragment and the local recombination rate to evaluate the performance without overfitting effects. Approximately 170,000 genomic DNA fragments of the human genome containing the Chromatin Immuno-Precipitation with high-throughput sequencing (ChIP-seq) peak of PRDM9 were used for constructing PWM and CNN. We found that CNN outperformed PWM in terms of area under the ROC curve and other metrics. Furthermore, the prediction scores of CNN correlated more strongly with the local recombination rate than PWM. We discuss that the superior performance of CNN would be in part due to the ability of CNN to capture the feature of surrounding sequences of actual PRDM9-binding sites.

show abstract

A Review About Transcription Factor Binding Sites Prediction Based on Deep Learning

Cited by 24 publications

References 104 publications

Deciphering transcription factors and their corresponding regulatory elements during inhibitory interneuron differentiation using deep neural networks

Deciphering transcription factors and their corresponding regulatory elements during inhibitory interneuron differentiation using deep neural networks

CBLANE: A deep learning approach for Transcription Factor Binding Sites Prediction

Predicting PRDM9 Binding Sites by a Convolutional Neural Network and Verification Using Genetic Recombination Map

Contact Info

Product

Resources

About