PreDBP-PLMs: Prediction of DNA-binding proteins based on pre-trained protein language models and convolutional neural networks.

The recognition of DNA-binding proteins (DBPs) is the crucial step to understanding their roles in various biological processes such as genetic regulation, gene expression, cell cycle control, DNA repair, and replication within cells. However, conventional experimental methods for identifying DBPs are usually time-consuming and expensive. Therefore, there is an urgent need to develop rapid and efficient computational methods for the prediction of DBPs. In this study, we proposed a novel predictor named PreDBP-PLMs to further improve the identification accuracy of DBPs by fusing the pre-trained protein language model (PLM) ProtT5 embedding with evolutionary features as input to the classic convolutional neural network (CNN) model. Firstly, the ProtT5 embedding was combined with different evolutionary features derived from the position-specific scoring matrix (PSSM) to represent protein sequences. Then, the optimal feature combination was selected and input to the CNN classifier for the prediction of DBPs. Finally, the 5-fold cross-validation (CV), the leave-one-out CV (LOOCV), and the independent set test were adopted to examine the performance of PreDBP-PLMs on the benchmark datasets. Compared to the existing state-of-the-art predictors, PreDBP-PLMs exhibits an accuracy improvement of 0.5% and 5.2% on the PDB186 and PDB2272 datasets, respectively. It demonstrated that the proposed method could serve as a useful tool for the recognition of DBPs.

Dissecting the structural and functional consequences of the evolutionary proline-glycine deletion in the wing 1 region of the forkhead domain of human FoxP1.

The human FoxP transcription factors dimerize via three-dimensional domain swapping, a unique feature among the human Fox family, as result of evolutionary sequence adaptations in the forkhead domain. This is the case for the conserved glycine and proline residues in the wing 1 region, which are absent in FoxP proteins but present in most of the Fox family. In this work, we engineered both glycine (G) and proline-glycine (PG) insertion mutants to evaluate the deletion events in FoxP proteins in their dimerization, stability, flexibility, and DNA-binding ability. We show that the PG insertion only increases protein stability, whereas the single glycine insertion decreases the association rate and protein stability and promotes affinity to the DNA ligand.

Comprehensive Analysis of the Expression, Prognosis, and Immune Infiltrates for Chromodomain-Helicase-DNA-Binding Proteins in Breast Tumor.

BACKGROUND: Several recent studies suggest that chromodomain-helicase -DNA-binding domains (CHDs) are linked with cancers. We explored the association between chromodomain-Helicase-DNA-binding domain proteins and breast cancer (BrCa) and introduced potential prognostic markers using various databases. MATERIALS AND METHODS: We analyzed the expression of the CHD family and their prognostic value in BrCa by mining UALCAN, TIMER, and Kaplan-Meier plotter databases. The association of CHD expression and immune infiltrating abundance was studied via the TIMER database. In addition, microRNAs related to the CHD family were identified by using the MirTarBase online database. RESULTS: The present study indicated that compared to normal tissues, BrCa tissues showed increased mRNA levels of CHD3/4/7 but decreased CHD2/5/9 expression. Interestingly, We also found a positive correlation between CHD gene expression and the infiltration of macrophage, neutrophil, and dendritic cells in BrCa, except CHD3/5. The Kaplan-Meier Plotter analysis suggested that high expression levels of CHD1/2/3/4/6/8/9 were significantly related to shorter relapse-free survival (RFS), while higher mRNA expression of CHD1, CHD2, CHD8, and CHD9 was significantly associated with longer overall survival of BrCa patients. The miRNAs of hsa-miR-615-3p and hsa-let-7b-5p were identified as being more correlated with the CHD family. CONCLUSION: The altered expression of some CHD members was significantly related to clinical cancer outcomes, and CHD1/2/8/9 could serve as potential prognostic biomarkers to improve the survival of BrCa patients. However, to evaluate the studied CHD members in detail are needed further investigations including experimental validation.

ProkDBP: Toward more precise identification of prokaryotic DNA binding proteins.

Prokaryotic DNA binding proteins (DBPs) play pivotal roles in governing gene regulation, DNA replication, and various cellular functions. Accurate computational models for predicting prokaryotic DBPs hold immense promise in accelerating the discovery of novel proteins, fostering a deeper understanding of prokaryotic biology, and facilitating the development of therapeutics targeting for potential disease interventions. However, existing generic prediction models often exhibit lower accuracy in predicting prokaryotic DBPs. To address this gap, we introduce ProkDBP, a novel machine learning-driven computational model for prediction of prokaryotic DBPs. For prediction, a total of nine shallow learning algorithms and five deep learning models were utilized, with the shallow learning models demonstrating higher performance metrics compared to their deep learning counterparts. The light gradient boosting machine (LGBM), coupled with evolutionarily significant features selected via random forest variable importance measure (RF-VIM) yielded the highest five-fold cross-validation accuracy. The model achieved the highest auROC (0.9534) and auPRC (0.9575) among the 14 machine learning models evaluated. Additionally, ProkDBP demonstrated substantial performance with an independent dataset, exhibiting higher values of auROC (0.9332) and auPRC (0.9371). Notably, when benchmarked against several cutting-edge existing models, ProkDBP showcased superior predictive accuracy. Furthermore, to promote accessibility and usability, ProkDBP (https://iasri-sg.icar.gov.in/prokdbp/) is available as an online prediction tool, enabling free access to interested users. This tool stands as a significant contribution, enhancing the repertoire of resources for accurate and efficient prediction of prokaryotic DBPs.

The intracellular visualization of exogenous DNA in fluorescence microscopy.

In the development of non-viral gene delivery vectors, it is essential to reliably localize and quantify transfected DNA inside the cell. To track DNA, fluorescence microscopy methods are commonly applied. These mostly rely on fluorescently labeled DNA, DNA binding proteins fused to a fluorescent protein, or fluorescence in situ hybridization (FISH). In addition, co-stainings are often used to determine the colocalization of the DNA in specific cellular compartments, such as the endolysosomes or the nucleus. We provide an overview of these DNA tracking methods, advice on how they should be combined, and indicate which co-stainings or additional methods are required to draw precise conclusions from a DNA tracking experiment. Some emphasis is given to the localization of exogenous DNA inside the nucleus, which is the last step of DNA delivery. We argue that suitable tools which allow for the nuclear detection of faint signals are still missing, hampering the rational development of more efficient non-viral transfection systems.

A Directed Evolution Protocol for Engineering Minimal Transcription Factors, Based on CIS Display.

Directed evolution is an efficient strategy for obtaining desired biomolecules. Since the 1990s, the emergence of display techniques has enabled high-throughput screening of functional proteins. However, classical methods require library construction by plasmid cloning and are limited by transformation efficiencies, typically limiting library sizes to ~106-107 variants. More recently, in vitro techniques have emerged that avoid cloning, allowing library sizes of >1012 members. One of these, CIS display, is a DNA-based display technique which allows high-throughput selection of biomolecules in vitro. CIS display creates the genotype-phenotype link required for selection by a DNA replication initiator protein, RepA, that binds exclusively to the template from which it has been expressed. This method has been successfully used to evolve new protein-protein interactions but has not been used before to select DNA-binding proteins, which are major components in mammalian synthetic biology. In this chapter, we describe a directed evolution method using CIS display to efficiently select functional DNA-binding proteins from pools of nonbinding proteins. The method is illustrated by enriching the minimal transcription factor Cro from a low starting frequency (1 in 109). This protocol is also applicable to engineering other DNA-binding proteins or transcription factors from combinatorial libraries.

Effect of TraN key residues involved in DNA binding on pIP501 transfer rates in Enterococcus faecalis.

Conjugation is a major mechanism that facilitates the exchange of antibiotic resistance genes among bacteria. The broad-host-range Inc18 plasmid pIP501 harbors 15 genes that encode for a type IV secretion system (T4SS). It is a membrane-spanning multiprotein complex formed between conjugating donor and recipient cells. The penultimate gene of the pIP501 operon encodes for the cytosolic monomeric protein TraN. This acts as a transcriptional regulator by binding upstream of the operon promotor, partially overlapping with the origin of transfer. Additionally, TraN regulates traN and traO expression by binding upstream of the PtraNO promoter. This study investigates the impact of nine TraN amino acids involved in binding to pIP501 DNA through site-directed mutagenesis by exchanging one to three residues by alanine. For three traN variants, complementation of the pIP501∆traN knockout resulted in an increase of the transfer rate by more than 1.5 orders of magnitude compared to complementation of the mutant with native traN. Microscale thermophoresis (MST) was used to assess the binding affinities of three TraN double-substituted variants and one triple-substituted variant to its cognate pIP501 double-stranded DNA. The MST data strongly correlated with the transfer rates obtained by biparental mating assays in Enterococcus faecalis. The TraN variants TraN_R23A-N24A-Q28A, TraN_H82A-R86A, and TraN_G100A-K101A not only exhibited significantly lower DNA binding affinities but also, upon complementation of the pIP501∆traN knockout, resulted in the highest pIP501 transfer rates. This confirms the important role of the TraN residues R23, N24, Q28, H82, R86, G100, and K101 in downregulating pIP501 transfer. Although TraN is not part of the mating pair formation complex, TraE, TraF, TraH, TraJ, TraK, and TraM were coeluted with TraN in a pull-down. Moreover, TraN homologs are present not only in Inc18 plasmids but also in RepA_N and Rep_3 family plasmids, which are frequently found in enterococci, streptococci, and staphylococci. This points to a widespread role of this repressor in conjugative plasmid transfer among Firmicutes.

Generation of site-specifically labelled fluorescent human XPA to investigate DNA binding dynamics during nucleotide excision repair.

Nucleotide excision repair (NER) promotes genomic integrity by removing bulky DNA adducts introduced by external factors such as ultraviolet light. Defects in NER enzymes are associated with pathological conditions such as Xeroderma Pigmentosum, trichothiodystrophy, and Cockayne syndrome. A critical step in NER is the binding of the Xeroderma Pigmentosum group A protein (XPA) to the ss/ds DNA junction. To better capture the dynamics of XPA interactions with DNA during NER we have utilized the fluorescence enhancement through non-canonical amino acids (FEncAA) approach. 4-azido-L-phenylalanine (4AZP or pAzF) was incorporated at Arg-158 in human XPA and conjugated to Cy3 using strain-promoted azide-alkyne cycloaddition. The resulting fluorescent XPA protein (XPACy3) shows no loss in DNA binding activity and generates a robust change in fluorescence upon binding to DNA. Here we describe methods to generate XPACy3 and detail in vitro experimental conditions required to stably maintain the protein during biochemical and biophysical studies.

DRBpred: A sequence-based machine learning method to effectively predict DNA- and RNA-binding residues.

DNA-binding and RNA-binding proteins are essential to an organism's normal life cycle. These proteins have diverse functions in various biological processes. DNA-binding proteins are crucial for DNA replication, transcription, repair, packaging, and gene expression. Likewise, RNA-binding proteins are essential for the post-transcriptional control of RNAs and RNA metabolism. Identifying DNA- and RNA-binding residue is essential for biological research and understanding the pathogenesis of many diseases. However, most DNA-binding and RNA-binding proteins still need to be discovered. This research explored various properties of the protein sequences, such as amino acid composition type, Position-Specific Scoring Matrix (PSSM) values of amino acids, Hidden Markov model (HMM) profiles, physiochemical properties, structural properties, torsion angles, and disorder regions. We utilized a sliding window technique to extract more information from a target residue's neighbors. We proposed an optimized Light Gradient Boosting Machine (LightGBM) method, named DRBpred, to predict DNA-binding and RNA-binding residues from the protein sequence. DRBpred shows an improvement of 112.00 %, 33.33 %, and 6.49 % for the DNA-binding test set compared to the state-of-the-art method. It shows an improvement of 112.50 %, 16.67 %, and 7.46 % for the RNA-binding test set regarding Sensitivity, Mathews Correlation Coefficient (MCC), and AUC metric.

Role of LL-37 in Oral Bacterial DNA Accumulation in Dental Plaque.

Dental plaque, a highly structured polymicrobial biofilm, persistently forms in the oral cavity and is a common problem affecting oral health. The role of oral defense factors in either collaborating or disrupting host-microbiome interactions remains insufficiently elucidated. This study aims to explore the role of LL-37, a critical antimicrobial peptide in the oral cavity, in dental plaque formation. Through immunostaining dental plaque specimens, we observed that LL-37 and DNA colocalized in the samples, appearing as condensed clusters. In vitro experiments revealed that LL-37 binds rapidly to oral bacterial DNA, forming high molecular weight, DNase-resistant complexes. This interaction results in LL-37 losing its inherent antibacterial activity. Further, upon the addition of LL-37, we observed a visible increase in the precipitation of bacterial DNA. We also discovered a significant correlation between the levels of the DNA-LL-37 complex and LL-37 within dental plaque specimens, demonstrating the ubiquity of the complex within the biofilm. By using immunostaining on dental plaque specimens, we could determine that the DNA-LL-37 complex was present as condensed clusters and small bacterial cell-like structures. This suggests that LL-37 immediately associates with the released bacterial DNA to form complexes that subsequently diffuse. We also demonstrated that the complexes exhibited similar Toll-like receptor 9-stimulating activities across different bacterial species, including Porphyromonas gingivalis, Fusobacterium nucleatum, Prevotella intermedia, and Streptococcus salivarius. However, these complexes prompted dissimilar activities, such as the production of IL-1ß in monocytic cells via both NLRP3 pathway-dependent and pathway-independent mechanisms. This study, therefore, reveals the adverse role of LL-37 in dental plaque, where it binds bacterial DNA to form complexes that may precipitate to behave like an extracellular matrix. Furthermore, the unveiled stimulating properties and species-dependent activities of the oral bacterial DNA-LL-37 complexes enrich our understanding of dental plaque pathogenicity and periodontal innate immune responses.

Double Stranded DNA Binding Stapled Peptides: An Emerging Tool for Transcriptional Regulation.

Stapled peptides have rapidly established themselves as a powerful technique to mimic α-helical interactions with a short peptide sequence. There are many examples of stapled peptides that successfully disrupt α-helix-mediated protein-protein interactions, with an example currently in clinical trials. DNA-protein interactions are also often mediated by α-helices and are involved in all transcriptional regulation processes. Unlike DNA-binding small molecules, which typically lack DNA sequence selectivity, DNA-binding proteins bind with high affinity and high selectivity. These are ideal candidates for the design DNA-binding stapled peptides. Despite the parallel to protein-protein interaction disrupting stapled peptides and the need for sequence specific DNA binders, there are very few DNA-binding stapled peptides. In this review we examine all the known DNA-binding stapled peptides. Their design concepts are compared to stapled peptides that disrupt protein-protein interactions and based on the few examples in the literature, DNA-binding stapled peptide trends are discussed.

LysR-type transcriptional regulators: state of the art.

The LysR-type transcriptional regulators (LTTRs) are DNA-binding proteins present in bacteria, archaea, and in algae. Knowledge about their distribution, abundance, evolution, structural organization, transcriptional regulation, fundamental roles in free life, pathogenesis, and bacteria-plant interaction has been generated. This review focuses on these aspects and provides a current picture of LTTR biology.

DBPMod: a supervised learning model for computational recognition of DNA-binding proteins in model organisms.

DNA-binding proteins (DBPs) play critical roles in many biological processes, including gene expression, DNA replication, recombination and repair. Understanding the molecular mechanisms underlying these processes depends on the precise identification of DBPs. In recent times, several computational methods have been developed to identify DBPs. However, because of the generic nature of the models, these models are unable to identify species-specific DBPs with higher accuracy. Therefore, a species-specific computational model is needed to predict species-specific DBPs. In this paper, we introduce the computational DBPMod method, which makes use of a machine learning approach to identify species-specific DBPs. For prediction, both shallow learning algorithms and deep learning models were used, with shallow learning models achieving higher accuracy. Additionally, the evolutionary features outperformed sequence-derived features in terms of accuracy. Five model organisms, including Caenorhabditis elegans, Drosophila melanogaster, Escherichia coli, Homo sapiens and Mus musculus, were used to assess the performance of DBPMod. Five-fold cross-validation and independent test set analyses were used to evaluate the prediction accuracy in terms of area under receiver operating characteristic curve (auROC) and area under precision-recall curve (auPRC), which was found to be ~89-92% and ~89-95%, respectively. The comparative results demonstrate that the DBPMod outperforms 12 current state-of-the-art computational approaches in identifying the DBPs for all five model organisms. We further developed the web server of DBPMod to make it easier for researchers to detect DBPs and is publicly available at https://iasri-sg.icar.gov.in/dbpmod/. DBPMod is expected to be an invaluable tool for discovering DBPs, supplementing the current experimental and computational methods.

PredDRBP-MLP: Prediction of DNA-binding proteins and RNA-binding proteins by multilayer perceptron.

Proteins interact with many molecules in order to maintain the vital activities in cells. Proteins that interact with DNA are called DNA-binding proteins (DBP), and proteins that interact with RNA are called RNA-binding proteins (RBP). Since DBPs and RBPs are involved in critical biological processes, their classification is quite important. Although the convolutional neural network and bidirectional long-short-term memory hybrid model (CNN-BiLSTM) is very popular in DBP and RBP classification, it has problems such as requirement of high processing power and long training time. Therefore, a multilayer perceptron (MLP) based predictor, PredDRBP-MLP (Predictor of DNA-Binding Proteins and RNA-Binding Proteins - Multilayer Perceptron) was developed in this study. PredDRBP-MLP is an artificial learning model that performs multi-class classification of DBPs, RBPs and non-nucleic acid-binding proteins (NNABP). PredDRBP-MLP achieved quite successful results on the independent dataset, specifically in the NNABP class, compared to the existing predictors, in addition to requiring lower processing power and being able to train quicker compared to CNN-BiLSTM based predictors. In NNABP class, PredDRBP-MLP predictor achieved 0.578 precision, 0.522 recall and 0.549 F1-score, while other multi-class predictor achieved 0.486 precision, 0.183 recall and 0.266 F1-score. A desktop application was developed for PredDRBP-MLP. The application is freely accessible at https://sourceforge.net/projects/preddrbp-mlp.

RNA-binding proteins as a common ground for neurodegeneration and inflammation in amyotrophic lateral sclerosis and multiple sclerosis.

The neurodegenerative and inflammatory illnesses of amyotrophic lateral sclerosis and multiple sclerosis were once thought to be completely distinct entities that did not share any remarkable features, but new research is beginning to reveal more information about their similarities and differences. Here, we review some of the pathophysiological features of both diseases and their experimental models: RNA-binding proteins, energy balance, protein transportation, and protein degradation at the molecular level. We make a thorough analysis on TDP-43 and hnRNP A1 dysfunction, as a possible common ground in both pathologies, establishing a potential link between neurodegeneration and pathological immunity. Furthermore, we highlight the putative variations that diverge from a common ground in an atemporal course that proposes three phases for all relevant molecular events.

A DNA-binding protein capture technology that purifies proteins by directly isolating the target DNA.

DNA-protein interactions are critical to almost all cellular functions, and identification of the proteins that bind to an DNA site of interest (gene-centered approach) is an important investigation area. However, gene-centered methods are mainly based on DNA hybridization to isolate target proteins, which is complex and inefficient. Here, we built a gene-centered approach involving direct isolation of target DNA, termed protein capture based on biolistic transformation (PCaB). The target DNA was labeled with biotin and cyanine 3 (Cy3) at its 5' and 3' DNA ends, respectively, and introduced into the host plants through biolistic transformation. The DNA and its binding proteins were crosslinked using formaldehyde. The labeled DNAs were obtained using gel excision and biotin-Streptavidin affinity according to the indication of Cy3 fluorescence, which make harvest of target DNA with a low background. The DNA-binding proteins were identified using mass spectrometry analysis. The PCaB method allowed us to identify and confirm 16 putative upstream regulators of the BpERF3 gene from Betula platyphylla. Theoretically, PCaB could be adapted to all plant species that can be transformed using biolistic bombardment, and captures DNA-binding proteins quickly with a low background. Therefore, PCaB will provide a powerful tool to discover DNA-protein interactions.

Identification of DNA-binding protein based multiple kernel model.

DNA-binding proteins (DBPs) play a critical role in the development of drugs for treating genetic diseases and in DNA biology research. It is essential for predicting DNA-binding proteins more accurately and efficiently. In this paper, a Laplacian Local Kernel Alignment-based Restricted Kernel Machine (LapLKA-RKM) is proposed to predict DBPs. In detail, we first extract features from the protein sequence using six methods. Second, the Radial Basis Function (RBF) kernel function is utilized to construct pre-defined kernel metrics. Then, these metrics are combined linearly by weights calculated by LapLKA. Finally, the fused kernel is input to RKM for training and prediction. Independent tests and leave-one-out cross-validation were used to validate the performance of our method on a small dataset and two large datasets. Importantly, we built an online platform to represent our model, which is now freely accessible via http://8.130.69.121:8082/.

BiCaps-DBP: Predicting DNA-binding proteins from protein sequences using Bi-LSTM and a 1D-capsule network.

Predicting DNA-binding proteins (DBPs) based solely on primary sequences is one of the most challenging problems in genome annotation. DBPs play a crucial role in various biological processes, including DNA replication, transcription, repair, and splicing. Some DBPs are essential in pharmaceutical research on various human cancers and autoimmune diseases. Existing experimental methods for identifying DBPs are time-consuming and costly. Therefore, developing a rapid and accurate computational technique is necessary to address the issue. This study introduces BiCaps-DBP, a deep learning-based method that improves DBP prediction performance by combining bidirectional long short-term memory with a 1D-capsule network. This study uses three training and independent datasets to evaluate the proposed model's generalizability and robustness. Based on three independent datasets, BiCaps-DBP achieved 1.05%, 5.79% and 0.40% higher accuracies than an existing predictor for PDB2272, PDB186 and PDB20000, respectively. These outcomes indicate that the proposed method is a promising DBP predictor.

Interactions between Fkh1 monomers stabilize its binding to DNA replication origins.

Eukaryotic DNA replication is initiated from multiple genomic origins, which can be broadly categorized as firing early or late in the S phase. Several factors can influence the temporal usage of origins to determine the timing of their firing. In budding yeast, the Forkhead family proteins Fkh1 and Fkh2 bind to a subset of replication origins and activate them at the beginning of the S phase. In these origins, the Fkh1/2 binding sites are arranged in a strict configuration, suggesting that Forkhead factors must bind the origins in a specific manner. To explore these binding mechanisms in more detail, we mapped the domains of Fkh1 that were required for its role in DNA replication regulation. We found that a short region of Fkh1 near its DNA binding domain was essential for the protein to bind and activate replication origins. Analysis of purified Fkh1 proteins revealed that this region mediates dimerization of Fkh1, suggesting that intramolecular contacts of Fkh1 are required for efficient binding and regulation of DNA replication origins. We also show that the Sld3-Sld7-Cdc45 complex is recruited to Forkhead-regulated origins already in the G1 phase and that Fkh1 is constantly required to keep these factors bound on origins before the onset of the S phase. Together, our results suggest that dimerization-mediated stabilization of DNA binding by Fkh1 is crucial for its ability to activate DNA replication origins.