Prediction Of Transcription Factor Binding Sites Research Articles

MLSNet: a deep learning model for predicting transcription factor binding sites.

Accurate prediction of transcription factor binding sites (TFBSs) is essential for understanding gene regulation mechanisms and the etiology of diseases. Despite numerous advances in deep learning for predicting TFBSs, their performance can still be enhanced. In this study, we propose MLSNet, a novel deep learning architecture designed specifically to predict TFBSs. MLSNet innovatively integrates multisize convolutional fusion with long short-term memory (LSTM) networks to effectively capture DNA-sparse higher-order sequence features. Further, MLSNet incorporates super token attention and Bi-LSTM to systematically extract and integrate higher-order DNA shape features. Experimental results on 165 ChIP-seq (chromatin immunoprecipitation followed by sequencing) datasets indicate that MLSNet consistently outperforms several state-of-the-art algorithms in the prediction of TFBSs. Specifically, MLSNet reports average metrics: 0.8306 for ACC, 0.8992 for AUROC, and 0.9035 for AUPRC, surpassing the second-best methods by 1.82%, 1.68%, and 1.54%, respectively. This research delineates the effectiveness of combining multi-size convolutional layers with LSTM and DNA shape-based features in enhancing predictive accuracy. Moreover, this study comprehensively assesses the variability in model performance across different cell lines and transcription factors. The source code of MLSNet is available at https://github.com/minghaidea/MLSNet.

Genetic association of GJA8 with long-segment Hirschsprung's disease in southern Chinese children.

Hirschsprung's disease (HSCR) is a complex congenital neurodevelopmental disorder affecting colons caused by both genetic and environmental factors. Although several genes have been identified as contributing factors in HSCR, the pathogenesis is still largely unclear, especially for the low prevalent long-segment HSCR (L-HSCR). Gap junction protein alpha 8 (GJA8) is involved in several physiological processes and has been implicated in several diseases. However, the relationship between GJA8 single nucleotide polymorphism (SNP) rs17160783 and HSCR in the southern Chinese population remains unknown. The study aimed to explore the association of genetic variants in GJA8 and HSCR susceptibility in southern Chinese. SNP rs17160783 A>G in GJA8 was genotyped by TaqMan SNP Genotyping Assay in all samples, which included 1,329 HSCR children (cases) and 1,473 healthy children (controls). Odds ratio (OR) and 95% confidence interval (CI) were used to evaluate the association of GJA8 polymorphisms with HSCR susceptibility. The GTEx database and transcription factor binding site (TFBS) prediction were used to analyze the potential regulatory function of rs17160783. Genetic association analysis illustrated that rs17160783 could increase the risk of L-HSCR (Padj=0.04, ORadj =1.48, 95% CI: 1.02-2.14). We also found that GJA8 expression was increased in HSCR and neurodevelopmentally impaired animal models. External epigenetic data revealed that GJA8 rs17160783 may have the potential to regulate the expression of the GJA8, possibly by altering the binding of transcription factors for GJA8, and consequently impacting the PI3K-Akt signaling pathway during the enteric nervous system (ENS) development. Our results suggested that rs17160783 might play a regulatory role in GJA8 expression and increase the susceptibility of L-HSCR in children from southern China.

Predicting transcription factor binding sites by a multi-modal representation learning method based on cross-attention network

The prediction of transcription factor binding sites (TFBS) plays a crucial role in studying cellular functions and understanding transcriptional regulatory processes. With the development of chromatin immunoprecipitation sequencing (ChIP-seq) technology, an increasing number of computer-aided TFBS prediction models have emerged. However, how to integrate multi-modal information of DNA and obtain efficient features to improve prediction accuracy remains a major challenge. Here, we propose MultiTF, a multi-modal representation learning method based on a cross-attention network for predicting transcription factor binding sites. Among TFBS prediction methods, we are the first to use graph neural networks and cross-attention networks for representation learning. MultiTF uses dna2vec to extract global contextual features of DNA sequences, DNAshapeR to extract shape features, and the CDPfold model and graph attention network for learning and representation of DNA structural features. Finally, with the help of our cross-attention module, we successfully combine sequence, structural, and shape features to achieve interactive fusion. When comparing MultiTF to other state-of-the-art methods using 165 ENCODE ChIP-seq datasets, we find that MultiTF exhibits average ACC, ROC-AUC, and PR-AUC values of 0.911, 0.978, and 0.982, respectively. The results show that MultiTF achieves unprecedented prediction accuracy compared to previous TFBS prediction models. In addition, our visual analysis of structural features provides interpretability for the prediction results.

BertSNR: an interpretable deep learning framework for single-nucleotide resolution identification of transcription factor binding sites based on DNA language model.

Transcription factors are pivotal in the regulation of gene expression, and accurate identification of transcription factor binding sites (TFBSs) at high resolution is crucial for understanding the mechanisms underlying gene regulation. The task of identifying TFBSs from DNA sequences is a significant challenge in the field of computational biology today. To address this challenge, a variety of computational approaches have been developed. However, these methods face limitations in their ability to achieve high-resolution identification and often lack interpretability. We propose BertSNR, an interpretable deep learning framework for identifying TFBSs at single-nucleotide resolution. BertSNR integrates sequence-level and token-level information by multi-task learning based on pre-trained DNA language models. Benchmarking comparisons show that our BertSNR outperforms the existing state-of-the-art methods in TFBS predictions. Importantly, we enhanced the interpretability of the model through attentional weight visualization and motif analysis, and discovered the subtle relationship between attention weight and motif. Moreover, BertSNR effectively identifies TFBSs in promoter regions, facilitating the study of intricate gene regulation. The BertSNR source code can be found at https://github.com/lhy0322/BertSNR.

Advancing DNA Language Models through Motif-Oriented Pre-Training with MoDNA

Acquiring meaningful representations of gene expression is essential for the accurate prediction of downstream regulatory tasks, such as identifying promoters and transcription factor binding sites. However, the current dependency on supervised learning, constrained by the limited availability of labeled genomic data, impedes the ability to develop robust predictive models with broad generalization capabilities. In response, recent advancements have pivoted towards the application of self-supervised training for DNA sequence modeling, enabling the adaptation of pre-trained genomic representations to a variety of downstream tasks. Departing from the straightforward application of masked language learning techniques to DNA sequences, approaches such as MoDNA enrich genome language modeling with prior biological knowledge. In this study, we advance DNA language models by utilizing the Motif-oriented DNA (MoDNA) pre-training framework, which is established for self-supervised learning at the pre-training stage and is flexible enough for application across different downstream tasks. MoDNA distinguishes itself by efficiently learning semantic-level genomic representations from an extensive corpus of unlabeled genome data, offering a significant improvement in computational efficiency over previous approaches. The framework is pre-trained on a comprehensive human genome dataset and fine-tuned for targeted downstream tasks. Our enhanced analysis and evaluation in promoter prediction and transcription factor binding site prediction have further validated MoDNA’s exceptional capabilities, emphasizing its contribution to advancements in genomic predictive modeling.

Vernalization promotes bolting in sugar beet by inhibiting the transcriptional repressors of BvGI.

Sugar beet (Beta vulgaris L.), a biennial sugar crop, contributes about 16% of the world's sugar production. The transition from vegetative growth, during which sugar accumulated in beet, to reproductive growth, during which sugar exhausted in beet, is determined by vernalization and photoperiod. GIGANTEA (GI) is a key photoperiodic flowering gene that is induced by vernalization in sugar beet. To identify the upstream regulatory factors of BvGI, candidate transcription factors (TF) that were co-expressed with BvGI and could bind to the BvGI promoter were screened based on weighted gene co-expression network analysis (WGCNA) and TF binding site prediction. Subsequently, their transcriptional regulatory role on the BvGI was validated through subcellular localization, dual-luciferase assays and yeast transformation tests. A total of 7,586 differentially expressed genes were identified after vernalization and divided into 18 co-expression modules by WGCNA, of which one (MEcyan) and two (MEdarkorange2 and MEmidnightblue) modules were positively and negatively correlated with the expression of BvGI, respectively. TF binding site predictions using PlantTFDB enabled the screening of BvLHY, BvTCP4 and BvCRF4 as candidate TFs that negatively regulated the expression of BvGI by affecting its transcription. Subcellular localization showed that BvLHY, BvTCP4 and BvCRF4 were localized to the nucleus. The results of dual-luciferase assays and yeast transformation tests showed that the relative luciferase activity and expression of HIS3 was reduced in the BvLHY, BvTCP4 and BvCRF4 transformants, which suggested that the three TFs inhibited the BvGI promoter. In addition, real-time quantitative reverse transcription PCR showed that BvLHY and BvTCP4 exhibited rhythmic expression characteristics similar to that of BvGI, while BvCRF4 did not. Our results revealed that vernalization crosstalked with the photoperiod pathway to initiate bolting in sugar beet by inhibiting the transcriptional repressors of BvGI.

MulTFBS: A Spatial-Temporal Network with Multichannels for Predicting Transcription Factor Binding Sites.

Revealing the mechanisms that influence transcription factor binding specificity is the key to understanding gene regulation. In previous studies, DNA double helix structure and one-hot embedding have been used successfully to design computational methods for predicting transcription factor binding sites (TFBSs). However, DNA sequence as a kind of biological language, the method of word embedding representation in natural language processing, has not been considered properly in TFBS prediction models. In our work, we integrate different types of features of DNA sequence to design a multichanneled deep learning framework, namely MulTFBS, in which independent one-hot encoding, word embedding encoding, which can incorporate contextual information and extract the global features of the sequences, and double helix three-dimensional structural features have been trained in different channels. To extract sequence high-level information effectively, in our deep learning framework, we select the spatial-temporal network by combining convolutional neural networks and bidirectional long short-term memory networks with attention mechanism. Compared with six state-of-the-art methods on 66 universal protein-binding microarray data sets of different transcription factors, MulTFBS performs best on all data sets in the regression tasks, with the average R2 of 0.698 and the average PCC of 0.833, which are 5.4% and 3.2% higher, respectively, than the suboptimal method CRPTS. In addition, we evaluate the classification performance of MulTFBS for distinguishing bound or unbound regions on TF ChIP-seq data. The results show that our framework also performs well in the TFBS classification tasks.

Predicting Transcription Factor Binding Sites with Deep Learning.

Prediction of binding sites for transcription factors is important to understand how the latter regulate gene expression and how this regulation can be modulated for therapeutic purposes. A consistent number of references address this issue with different approaches, Machine Learning being one of the most successful. Nevertheless, we note that many such approaches fail to propose a robust and meaningful method to embed the genetic data under analysis. We try to overcome this problem by proposing a bidirectional transformer-based encoder, empowered by bidirectional long-short term memory layers and with a capsule layer responsible for the final prediction. To evaluate the efficiency of the proposed approach, we use benchmark ChIP-seq datasets of five cell lines available in the ENCODE repository (A549, GM12878, Hep-G2, H1-hESC, and Hela). The results show that the proposed method can predict TFBS within the five different cell lines very well; moreover, cross-cell predictions provide satisfactory results as well. Experiments conducted across cell lines are reinforced by the analysis of five additional lines used only to test the model trained using the others. The results confirm that prediction across cell lines remains very high, allowing an extensive cross-transcription factor analysis to be performed from which several indications of interest for molecular biology may be drawn.

Deep learning model for predicting genetic diseases using DNA sequence data

The emergence of machine learning in the recent decade has excelled in determining new potential features and nonlinear relationships existing between the data derived from the DNA sequences of genetic diseases. Machine learning also enhances the process of handling data with maximum predicted variables compared to observations during the data mining process of prediction. In this context, our study presents a deep learning model for predicting Transcription Factor Binding Sites (TFBS) in DNA sequences, with a focus on features within genetic data associated with diseases. Transcription Factors (TFs) play a crucial role in modulating gene expression by binding to TFBS. The accurate prediction of TFBS is essential for understanding genome function and evolution. Thus, we develop an efficient deep learning model that considers TFBS prediction as a nucleotide-level binary classification task. In our proposed model, first we create an input matrix using the original DNA sequences. Next, we encode these DNA sequences using one-hot encoding, representing them as a sequence of numerical values. We then employ three convolutional layers, allowing our model to capture intricate patterns and motif features over a larger spatial range. To capture important features within the DNA sequence and to focus on them, we incorporate an attention layer. Finally, a dense layer, consisting of two fully connected layers and a dropout layer, calculates the probability of TF binding site occurrence based on the features learned by the proposed model. Our experimental results, using in-vivo datasets obtained from Chip-seq, demonstrate the superior performance of our proposed deep learning model in TFBS prediction compared to other existing state-of-the-art methods. The improvement in accuracy is due to additional layers of CNN and then an attention layer in the model. Thus, this result in a better performance of our approach in predicting the transcription factor binding sites and enhancing our understanding of gene regulation and genome function.

TBCA: Prediction of transcription factor binding sites using a deep neural network with lightweight attention mechanism.

The identification of transcription factor binding sites (TFBSs) is crucial for understanding the regulatory mechanisms of gene expression, which contributes to unraveling cellular functions and disease development. Currently, the most common approach involves the use of deep learning techniques to predict TFBSs by combining sequence and shape features. Although significant progress has been made with these methods, the integration of local features extracted from DNA sequences and shapes with global features has not yet reached a sufficient level, and there is still significant room for improvement in the accuracy of prediction results. In this paper, we propose a novel framework based on convolution and attention mechanisms, referred to as TBCA, which combines DNA sequence information and shape information for predicting transcription factor binding sites. In this work, we employ a two-layer convolutional neural network (CNNs) and self-attention mechanism to extract complex sequence features from DNA. What's more, we utilize a Fourier-transform-enhanced multi-head attention along with channel attention to extract high-order shape features of DNA. Finally, these high-order sequence and shape features are integrated into the channel dimension to achieve accurate TFBSs prediction. Our research results demonstrate that TBCA exhibits superior predictive performance in 165 validated ChIP-seq datasets. Furthermore, the employed attention mechanisms can automatically learn important features at different positions and scales, enhancing the accuracy and robustness of feature representation. We also conduct an in-depth analysis of the contributions of five different shapes to site prediction, revealing that shape features can enhance the prediction of transcription factor DNA binding.

GNNMF: a multi-view graph neural network for ATAC-seq motif finding

BackgroundThe Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) utilizes the Transposase Tn5 to probe open chromatic, which simultaneously reveals multiple transcription factor binding sites (TFBSs) compared to traditional technologies. Deep learning (DL) technology, including convolutional neural networks (CNNs), has successfully found motifs from ATAC-seq data. Due to the limitation of the width of convolutional kernels, the existing models only find motifs with fixed lengths. A Graph neural network (GNN) can work on non-Euclidean data, which has the potential to find ATAC-seq motifs with different lengths. However, the existing GNN models ignored the relationships among ATAC-seq sequences, and their parameter settings should be improved.ResultsIn this study, we proposed a novel GNN model named GNNMF to find ATAC-seq motifs via GNN and background coexisting probability. Our experiment has been conducted on 200 human datasets and 80 mouse datasets, demonstrated that GNNMF has improved the area of eight metrics radar scores of 4.92% and 6.81% respectively, and found more motifs than did the existing models.ConclusionsIn this study, we developed a novel model named GNNMF for finding multiple ATAC-seq motifs. GNNMF built a multi-view heterogeneous graph by using ATAC-seq sequences, and utilized background coexisting probability and the iterloss to find different lengths of ATAC-seq motifs and optimize the parameter sets. Compared to existing models, GNNMF achieved the best performance on TFBS prediction and ATAC-seq motif finding, which demonstrates that our improvement is available for ATAC-seq motif finding.

Multiomics-integrated deep language model enables in silico genome-wide detection of transcription factor binding site in unexplored biosamples.

Transcription factor binding sites (TFBS) are regulatory elements that have significant impact on transcription regulation and cell fate determination. Canonical motifs, biological experiments, and computational methods have made it possible to discover TFBS. However, most existing in silico TFBS prediction models are solely DNA-based, and are trained and utilized within the same biosample, which fail to infer TFBS in experimentally unexplored biosamples. Here, we propose TFBS prediction by modified TransFormer (TFTF), a multimodal deep language architecture which integrates multiomics information in epigenetic studies. In comparison to existing computational techniques, TFTF has state-of-the-art accuracy, and is also the first approach to accurately perform genome-wide detection for cell-type and species-specific TFBS in experimentally unexplored biosamples. Compared to peak calling methods, TFTF consistently discovers true TFBS in threshold tuning-free way, with higher recalled rates. The underlying mechanism of TFTF reveals greater attention to the targeted TF's motif region in TFBS, and general attention to the entire peak region in non-TFBS. TFTF can benefit from the integration of broader and more diverse data for improvement and can be applied to multiple epigenetic scenarios. We provide a web server (https://tftf.ibreed.cn/) for users to utilize TFTF model. Users can train TFTF model and discover TFBS with their own data.

Reporting Two Rare Mutations With Associated Clinical Features in Bangladeshi Hb E/B Thalassemia Patients

Thalassemia is one of the most common genetic disorders in Bangladesh. We have performed an investigation among 360 patients with Hb E/β-thalassemia to understand the heterogeneity in thalassemia severity among Bangladeshi patients. The analysis revealed that one of the common mutations CD26 (G>A) is found with two rare mutations HBB: c.-79A>G (-29A>G) and HBB: c.45_46insG (CD14/15 (+G)) in two patients individually which are not prevalent in geographically adjacent populations. Among these two rare mutations, -29A>G is located at the ‘TATA’ box region of the promoter sequence regulating transcription initiation and CD14/15 (+G) leads to a premature stop codon. We performed transcription factor binding site prediction analysis and found that ‘AP-1’, ‘c-Jun’, and ‘kr’ transcription factors bind at the ‘TATA’ box region. From this analysis, we can predict that -29A>G mutation can alter the level of HBB transcription whereas the CD14/15 (+G) mutation causes a truncated (and possibly non-functional) HBB protein. Aligned with such assumption, it was observed from patients’ medical history that, the patient who had CD14/15 (+G) and CD26 (G>A) mutations was severely blood transfusion-dependent whereas the patient carrying -29A>G and CD26 (G>A) mutations was moderately transfusion dependent. The comprehensive analysis of the mutations offers crucial insights and approaches for enhanced management of thalassemia among the Bangladeshi population. Bioresearch Commu. 10(1): 1442-1448, 2024 (January)

Identification and characterization of the cupin_1 domain-containing proteins in ma bamboo (Dendrocalamus latiflorus) and their potential role in rhizome sprouting.

Cupin_1 domain-containing protein (CDP) family, which is a member of the cupin superfamily with the most diverse functions in plants, has been found to be involved in hormone pathways that are closely related to rhizome sprouting (RS), a vital form of asexual reproduction in plants. Ma bamboo is a typical clumping bamboo, which mainly reproduces by RS. In this study, we identified and characterized 53 Dendrocalamus latiflorus CDP genes and divided them into seven subfamilies. Comparing the genetic structures among subfamilies showed a relatively conserved gene structure within each subfamily, and the number of cupin_1 domains affected the conservation among D. latiflorus CDP genes. Gene collinearity results showed that segmental duplication and tandem duplication both contributed to the expansion of D. latiflorus CDP genes, and lineage-specific gene duplication was an important factor influencing the evolution of CDP genes. Expression patterns showed that CDP genes generally had higher expression levels in germinating underground buds, indicating that they might play important roles in promoting shoot sprouting. Transcription factor binding site prediction and co-expression network analysis indicated that D. latiflorus CDPs were regulated by a large number of transcription factors, and collectively participated in rhizome buds and shoot development. This study significantly provided new insights into the evolutionary patterns and molecular functions of CDP genes, and laid a foundation for further studying the regulatory mechanisms of plant rhizome sprouting.

Characterizing genetic variation in the regulation of the ER stress response through computational and cis-eQTL analyses.

Misfolded proteins in the endoplasmic reticulum (ER) elicit the ER stress response, a large transcriptional response driven by 3 well-characterized transcription factors (TFs). This transcriptional response is variable across different genetic backgrounds. One mechanism in which genetic variation can lead to transcriptional variability in the ER stress response is through altered binding and activity of the 3 main TFs: XBP1, ATF6, and ATF4. This work attempts to better understand this mechanism by first creating a computational pipeline to identify potential binding sites throughout the human genome. We utilized GTEx data sets to identify cis-eQTLs that fall within predicted TF binding sites (TFBSs). We also utilized the ClinVar database to compare the number of pathogenic vs benign variants at different positions of the binding motifs. Finally, we performed a cis-eQTL analysis on human cell lines experiencing ER stress to identify cis-eQTLs that regulate the variable ER stress response. The majority of these cis-eQTLs are unique to a given condition: control or ER stress. Some of these stress-specific cis-eQTLs fall within putative binding sites of the 3 main ER stress response TFs, providing a potential mechanism by which these cis-eQTLs might be impacting gene expression under ER stress conditions through altered TF binding. This study represents the first cis-eQTL analysis on human samples experiencing ER stress and is a vital step toward identifying the genetic components responsible for the variable ER stress response.

Using Fully Convolutional Network to Locate Transcription Factor Binding Sites Based on DNA Sequence and Conservation Information.

Transcription factors (TFs) play a part in gene expression. TFs can form complex gene expression regulation system by combining with DNA. Thereby, identifying the binding regions has become an indispensable step for understanding the regulatory mechanism of gene expression. Due to the great achievements of applying deep learning (DL) to computer vision and language processing in recent years, many scholars are inspired to use these methods to predict TF binding sites (TFBSs), achieving extraordinary results. However, these methods mainly focus on whether DNA sequences include TFBSs. In this paper, we propose a fully convolutional network (FCN) coupled with refinement residual block (RRB) and global average pooling layer (GAPL), namely FCNARRB. Our model could classify binding sequences at nucleotide level by outputting dense label for input data. Experimental results on human ChIP-seq datasets show that the RRB and GAPL structures are very useful for improving model performance. Adding GAPL improves the performance by 9.32% and 7.61% in terms of IoU (Intersection of Union) and PRAUC (Area Under Curve of Precision and Recall), and adding RRB improves the performance by 7.40% and 4.64%, respectively. In addition, we find that conservation information can help locate TFBSs.

Genome-wide identification of the banana GLR gene family and its expression analysis in response to low temperature and abscisic acid/methyl jasmonate

Glutamate receptor-like (GLR) is an important class of Ca2+ channel proteins, playing important roles in plant growth and development as well as in response to biotic and abiotic stresses. In this paper, we performed genome-wide identification of banana GLR gene family based on banana genomic data. Moreover, we analyzed the basic physicochemical properties, gene structure, conserved motifs, promoter cis-acting elements, evolutionary relationships, and used real-time fluorescence quantitative polymerase chain reaction (RT-qPCR) to verify the expression patterns of some GLR family members under low temperature of 4 ℃ and different hormone treatments. The results showed that there were 19 MaGLR family members in Musa acuminata, 16 MbGLR family members in Musa balbisiana and 14 MiGLR family members in Musa itinerans. Most of the members were stable proteins and had signal peptides, all of them had 3-6 transmembrane structures. Prediction of subcellular localization indicated that all of them were localized on the plasma membrane and irregularly distributed on the chromosome. Phylogenetic analysis revealed that banana GLRs could be divided into 3 subclades. The results of promoter cis-acting elements and transcription factor binding site prediction showed that there were multiple hormone- and stress-related response elements and 18 TFBS in banana GLR. RT-qPCR analysis showed that MaGLR1.1 and MaGLR3.5 responded positively to low temperature stress and were significantly expressed in abscisic acid/methyl jasmonate treatments. In conclusion, the results of this study suggest that GLR, a highly conserved family of ion channels, may play an important role in the growth and development process and stress resistance of banana.

DeepSTF: predicting transcription factor binding sites by interpretable deep neural networks combining sequence and shape.

Precise targeting of transcription factor binding sites (TFBSs) is essential to comprehending transcriptional regulatory processes and investigating cellular function. Although several deep learning algorithms have been created to predict TFBSs, the models' intrinsic mechanisms and prediction results are difficult to explain. There is still room for improvement in prediction performance. We present DeepSTF, a unique deep-learning architecture for predicting TFBSs by integrating DNA sequence and shape profiles. We use the improved transformer encoder structure for the first time in the TFBSs prediction approach. DeepSTF extracts DNA higher-order sequence features using stacked convolutional neural networks(CNNs), whereas rich DNA shape profiles are extracted by combining improved transformer encoder structure and bidirectional long short-term memory(Bi-LSTM), and, finally, the derived higher-order sequence features and representative shape profiles are integrated into the channel dimension to achieve accurate TFBSs prediction. Experiments on 165 ENCODE chromatin immunoprecipitation sequencing (ChIP-seq) datasets show that DeepSTF considerably outperforms several state-of-the-art algorithms in predicting TFBSs, and we explain the usefulness of the transformer encoder structure and the combined strategy using sequence features and shape profiles in capturing multiple dependencies and learning essential features. In addition, this paper examines the significance of DNA shape features predicting TFBSs. The source code of DeepSTF is available at https://github.com/YuBinLab-QUST/DeepSTF/.

MachineTFBS: Motif-based method to predict transcription factor binding sites with first-best models from machine learning library

This study aims to model the high-affinity transcription factor (TF) binding sites (TFBSs) in Saccharomyces cerevisiae promoters using high-throughput experimental TF binding data to train, test, and validate models. We introduce MachineTFBS, a machine learning based TFBS prediction tool consisting of a library of machine learning models generated by treating each TF's protein binding microarray (PBM) dataset as an individual optimization problem and selectively generating the first-best performing models. Since the modeling quality of machine learning methods varies for different TFs, we tested Random Forest, eXtreme Gradient Boosting, and up to 5-depth Deep Learning models. We extracted 274 high-resolution PBM datasets of 159 S. cerevisiae TFs from the UniProbe database. We designed an algorithm for the greedy selection of the first-best models and feature combinations from five feature extraction functions for each PBM data. The optimal subset of the functions was identified using a feature elimination algorithm. The feature elimination algorithm advances in successive cycles for each PBM data until the elimination of any feature(s) improves model performance or the feature set size is reduced to 1; in turn, reports the optimal subset. MachineTFBS predicts TFBSs with an average Matthews Correlation Coefficient score of 0.873.

HAMPLE: deciphering TF-DNA binding mechanism in different cellular environments by characterizing higher-order nucleotide dependency.

Transcription factor (TF) binds to conservative DNA binding sites in different cellular environments and development stages by physical interaction with interdependent nucleotides. However, systematic computational characterization of the relationship between higher-order nucleotide dependency and TF-DNA binding mechanism in diverse cell types remains challenging. Here, we propose a novel multi-task learning framework HAMPLE to simultaneously predict TF binding sites (TFBS) in distinct cell types by characterizing higher-order nucleotide dependencies. Specifically, HAMPLE first represents a DNA sequence through three higher-order nucleotide dependencies, including k-mer encoding, DNA shape and histone modification. Then, HAMPLE employs the customized gate control and the channel attention convolutional architecture to further capture cell-type-specific and cell-type-shared DNA binding motifs and epigenomic languages. Finally, HAMPLE exploits the joint loss function to optimize the TFBS prediction for different cell types in an end-to-end manner. Extensive experimental results on seven datasets demonstrate that HAMPLE significantly outperforms the state-of-the-art approaches in terms of auROC. In addition, feature importance analysis illustrates that k-mer encoding, DNA shape and histone modification have predictive power for TF-DNA binding in different cellular environments and are complementary to each other. Furthermore, ablation study and interpretable analysis validate the effectiveness of the customized gate control and the channel attention convolutional architecture in characterizing higher-order nucleotide dependencies. The source code is available at https://github.com/ZhangLab312/Hample. Supplementary data are available at Bioinformatics online.