In-silico Approach Research Articles

Bioactive metabolites of licorice and thyme as potential inhibitors of Cox1 enzyme of phytopathogens of Capsicum annuum L.: In-silico approaches

Cytochrome c oxidase subunit 1 (Cox1), a key enzyme, has a crucial role in cellular respiration in eukaryotes and prokaryotes. Generally, respiratory inhibitors are considered one of the types of chemical pesticides. Thyme oil and licorice aqueous extract have been reported to have antifungal activities against fungal phytopathogens of Capsicum annuum L., i.e., Colletotrichum capsici, Fusarium oxysporum, and Pythium aphanidermatum. The present study focuses on identifying the key bioactive molecules of thyme and licorice botanicals inhibiting the activity of the Cox1 enzymes of the above mentioned phytopathogens, employing the in-silico approach. From a wide range of bioactive molecules screened, the molecular docking indicated trans-carveol, carvacrol, kaempferol 3-rhamnoside 7-xyloside, kaempferitrin, and astragalin 7-rhamnoside as the potential inhibitors for Cox1 of C. capsici, β-Caryophyllene, Caryophyllene acetate, hispaglabridin A, kaempferol 3-rhamnoside 7-xyloside and licorice glycoside A for Cox1 of F. oxysporum and (+)-Longifolen, Caryophyllene acetate, Hispaglabridin A, Neoliquiritin 2''-apioside and Licorice-saponin A3 for Cox1 of P. aphanidermatum. Most of the top-scoring bioactive molecules exhibited higher binding affinity with the targets than the chemical compound, i.e., carbendazim. Density functional theory (DFT) analysis confirmed the reactivity of the top-docked compounds. Molecular dynamic simulations confirmed the stability of docked complexes when evaluated through multiple descriptors. Additionally, MM/PBSA analysis supported the findings, indicating the spontaneous binding of the enzymes to the screened ligands. ADMET analysis revealed the safety of the selected bioactive compounds. The present findings could be useful in developing biopesticidal formulations as efficient and sustainable alternatives to chemical pesticides. Communicated by Ramaswamy H. Sarma

Elucidating the role of phytocompounds from Brassica oleracea var. italic (Broccoli) on hyperthyroidism: an in-silico approach.

The online version contains supplementary material available at 10.1007/s40203-023-00180-2.

Tubeimosides are pan-coronavirus and filovirus inhibitors that can block their fusion protein binding to Niemann-Pick C1.

SARS-CoV-2 and filovirus enter cells via the cell surface angiotensin-converting enzyme 2 (ACE2) or the late-endosome Niemann-Pick C1 (NPC1) as a receptor. Here, we screened 974 natural compounds and identified Tubeimosides I, II, and III as pan-coronavirus and filovirus entry inhibitors that target NPC1. Using in-silico, biochemical, and genomic approaches, we provide evidence that NPC1 also binds SARS-CoV-2 spike (S) protein on the receptor-binding domain (RBD), which is blocked by Tubeimosides. Importantly, NPC1 strongly promotes productive SARS-CoV-2 entry, which we propose is due to its influence on fusion in late endosomes. The Tubeimosides' antiviral activity and NPC1 function are further confirmed by infection with SARS-CoV-2 variants of concern (VOC), SARS-CoV, and MERS-CoV. Thus, NPC1 is a critical entry co-factor for highly pathogenic human coronaviruses (HCoVs) in the late endosomes, and Tubeimosides hold promise as a new countermeasure for these HCoVs and filoviruses.

Evolution of protein domain repertoires of CALHM6.

Calcium (Ca2+) homeostasis is essential in conducting various cellular processes including nerve transmission, muscular movement, and immune response. Changes in Ca2+concentration in the cytoplasm are significant in bringing about various immune responses such as pathogen clearance and apoptosis. Various key players are involved in calcium homeostasis such as calcium binders, pumps, and channels. Sequence-based evolutionary information has recently been exploited to predict the biophysical behaviors of proteins, giving critical clues about their functionality. Ion channels are reportedly the first channels developed during evolution. Calcium homeostasis modulator protein 6 (CALHM6) is one such channel. Comprised of a single domain called Ca_hom_mod, CALHM6 is a stable protein interacting with various other proteins in calcium regulation. No previous attempt has been made to trace the exact evolutionary events in the domain of CALHM6, leaving plenty of room for exploring its evolution across a wide range of organisms. The current study aims to answer the questions by employing a computational-based strategy that used profile Hidden Markov Models (HMMs) to scan for the CALHM6 domain, integrated the data with a time-calibrated phylogenetic tree using BEAST and Mesquite, and visualized through iTOL. Around 4,000 domains were identified, and 14,000 domain gain, loss, and duplication events were observed at the end which also included various protein domains other than CALHM6. The data were analyzed concerning CALHM6 evolution as well as the domain gain, loss, and duplication of its interacting partners: Calpain, Vinculin, protein S100-A7, Thioredoxin, Peroxiredoxin, and Calmodulin-like protein 5. Duplication events of CALHM6 near higher eukaryotes showed its increasing complexity in structure and function. This in-silico phylogenetic approach applied to trace the evolution of CALHM6 was an effective approach to get a better understanding of the protein CALHM6.

Identifying potential inhibitors of phosphatidylinositol 4,5-bisphosphate 3-kinase: Molecular dynamic insights into the interaction and inhibitory mechanism

Phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) plays a crucial role in signaling pathways and has emerged as an attractive target for anticancer therapy. Developing small-molecule inhibitors targeting PIK3CA has emerged as a promising therapeutic strategy. Here, we employed in-silico approaches to identify potential inhibitors of PIK3CA from a pool of bioactive phytoconstituents available from IMPPAT 2.0 database. The initial screening was based on their drug-like properties and docking score, followed by interaction analysis and 100 nanoseconds (ns) molecular dynamics (MD) simulations to evaluate their conformational stability with PIK3CA. MD simulation studies suggested the formation of stable complexes between PIK3CA and the elucidated compounds Apollinine and Isoglycyrol. These compounds exhibited exceptional drug-like properties, potential binding efficiency, and remarkable stability. The findings suggest that Apollinine and Isoglycyrol could serve as potential leads in the therapeutic development of PIK3CA inhibitors for targeting cancer. Moreover, the insights gained from the simulation dynamics shed light on the molecular mechanisms of PIK3CA inhibition that facilitate the rational design of future inhibitors.

Insilico Approach to Find Bioactive Compound against Fruit Rot Disease (FRD) in Areca Catechu

Phytophthora meadii is a fungal species which causes Fruit Rot Disease (FRD) in Areca catechu affects the quality and yield which is main threat for agricultural economy. Traditional methods include chemical fungicides like bordeaux mixture, Fosetyl-al, metalaxyl, mancozeb, copper oxychloride, cymoxalin, copper sulfate, etc in treatment of FRD, which can lead to environmental and health issues. In this study, an in silico approach was employed to identify potential bioactive compounds that could serve as effective agents against the pathogens responsible for FRD in Areca catechu. Utilizing a combination of molecular docking and virtual screening, we screened a comprehensive library of natural compounds for their ability to inhibit key enzymes and proteins associated with FRD pathogens. The molecular docking analysis focused on identifying compounds with high binding affinities to target proteins, suggesting their potential efficacy in disrupting pathogen activity. Among the screened compounds, several compounds have shown strong binding affinities indicating their potential as bioactive agents against FRD pathogens. These findings provide a promising foundation for the development of natural, environmentally-friendly treatments for FRD in Areca catechu.

Completing the pieces of a puzzle: in-depth probing of Toxoplasma gondii rhoptry protein 4 as a promising target for vaccination using an in-silico approach.

The present study aimed to evaluate the key characteristics of Toxoplasma gondii rhoptry protein 4 (TgROP4), including physicochemical parameters, structural features, immunogenic epitopes, and virtual immune simulation, using several bioinformatics-based servers and tools. Based on allergenicity and antigenicity outputs, the TgROP4 protein seemed to have an immunogenic and non-allergenic nature. The quality of the three-dimensional (3D) structure improved after refinement, according to the outcomes of the Ramachandran plot and the ProSA-web servers. ABCpred and SVMTriP web tools were used to predict linear B lymphocyte epitopes and found several promising epitopes. Acceptable antigenicity, hydrophilicity, beta-turn, Bepipred linear epitope 2.0, flexibility, and surface accessibility scores were obtained through the Immune Epitope Database (IEDB). Also, seven discontinuous B-cell epitopes ranging from scores 0.966 to 0.848 were found in the 3D model of TgROP4 via the ElliPro. The IEDB findings showed T-cell epitopes on TgROP4 protein are capable to strongly bind to the major histocompatibility complex classes. In silico immune simulation was performed using C-ImmSim server and showed three injections of TgROP4 protein at 4-week intervals is capable to elicit adequate humoral and cell-mediated immune responses.

Analysing the Functional and Structural Impact of Single Nucleotide Polymorphisms in Matrix Metalloproteinase 3 Gene: An In-silico Approach

Introduction: Matrix Metalloproteinase 3 (MMP3) is a vital member of the MMP family, known for its wide range of substrate specificity and proteolytic activity against Extracellular Matrix (ECM) components. The role of functional polymorphisms in the MMP genes has been previously investigated in relation to cancer susceptibility, particularly breast cancer. Several SingleNucleotide Polymorphisms (SNPs) in the MMP3 gene have been linked to a number of clinical illnesses, such as Coronary Artery Disease (CAD); however, the results were not entirely conclusive. Aim: To identify pathogenic missense SNPs in the human MMP3 gene and analyse their effects on structure and function. Materials and Methods: This was a record-based cross-sectional study performed using data retrieved from online resources. The analysis was conducted using a series of different bioinformatic tools, for which ethical clearance was obtained from the institution. The online tools used included Sorting Intolerant from Tolerant (SIFT), PolyPhen-2, PhD-SNP, PANTHER, PROVEAN, and SNPs and GO to predict harmful non synonymous SNPs (nsSNPs). Further analysis was performed using I-Mutant 2.0, MutPred2, Consurf, and HOPE software. These tools were able to filter out damaging SNPs and predict the impact of deleterious SNPs on the structure and function of the MMP3 protein. Results: This study predicted two potentially pathogenic SNPs (D175Y and Y116C) out of 443 missense SNPs from dbSNP, which is a database of SNPs available on the National Centre for Biotechnology Information website. Further analysis revealed that these SNPs were located in highly conserved regions and were predicted to decrease protein stability. Conclusion: In this study, two potentially pathogenic SNPs (D175Y and Y116C) were identified. Characterisation of these SNPs can help us gain a better understanding of the molecular basis of clinical conditions. The results of this study can be further validated by designing population-based studies and wet lab experiments. This will help in augmenting research and personalised medicine.

Andrographolide demonstrates anti-proliferative activity in oral cancer by promoting apoptosis, the programmed cell death process.

Andrographolide has been studied on different types of human cancer cells, but very few studies have been conducted on oral cancer. The study aimed to evaluate the anticancer potential of Andrographolide on an oral cancer cell line (KB) through in-silico network analysis and in vitro assays. The in-silico analysis involved the determination of drug-likeness prediction, prediction of common targets between oral cancer and andrographolide, Protein-Protein Interactions (PPI), hub genes, top 10 associated pathways by Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway, gene ontology (GO), and molecular docking experiments. In vitro assays comprised MTT assay, apoptosis assay, cell cycle analysis, intracellular reactive oxygen species (ROS) measurement, mitochondrial membrane potential (MMP), anti-migration activity, and gene expressions using polymerase chain reaction (PCR). Fifteen common genes were obtained and were seen to be involved in cellular proliferation, regulation of apoptosis, migration of cells, regulation of MAPK cascade, and regulation of cell cycle. The most common genes involved in the top 10 pathways were MAPK1, MAPK8, MAPK14, and IL6 which were seen to be associated with the MAPK signaling pathway which may be the key pathway through which andrographolide may aid in treating oral cancer. In vitro assays showed anti-proliferative properties, late apoptosis, and anti-migratory properties. According to the results obtained, andrographolide has shown anticancer properties and has the potential to be used as a chemotherapeutic drug. The in-silico approach used in the present study can aid as a model for future research in developing efficient cancer treatments.

An in-silico approach for the identification of natural compounds as potential BACE1 inhibitors for the treatment of Alzheimer disease

An in-silico approach for the identification of natural compounds as potential BACE1 inhibitors for the treatment of Alzheimer disease

In Silico Exploration of few TriazoloPyrimidine Derivatives as Virtual Inhibitors against SARS-CoV-2 : A Comprehensive Analysis Integrating Molecular Docking and ADME-Toxicity Evaluation

This ground-breaking research delves into the potential of novel triazolopyrimidine derivatives as inhibitors of SARS-CoV, employing a robust in-silico approach that integrates molecular docking studies, molecular dynamics simulations, and comprehensive ADMET parameter assessments. The overarching goal is to identify compounds that exhibit both robust binding affinity and favourable pharmacokinetic profiles, laying the foundation for potential antiviral drug development. The methodology begins with the meticulous selection of triazolo pyrimidine derivatives based on their structural characteristics. Molecular docking studies are then conducted, focusing on a specific binding site on the SARS-CoV target. Subsequent molecular dynamics simulations provide a dynamic perspective on the stability of the binding interactions over time. ADMET parameter assessments are employed to evaluate drug-likeness and safety, crucial factors in determining the compounds' viability as potential drug candidates. The findings of this research underscore promising interactions between the triazolo pyrimidine derivatives and the targeted viral site, suggesting their potential as inhibitors of SARS-CoV. Importantly, the ADMET assessments contribute valuable insights into the pharmacokinetic properties of the compounds, informing their overall safety and suitability for further development. Notably, molecular docking studies include a comparative analysis with the standard antiviral drug remdesivir, revealing that the triazolo pyrimidine derivatives exhibit energy scores surpassing those of remdesivir. This finding signifies a potential advantage in terms of binding affinity and effectiveness against SARS-CoV. The implications of this research are profound, extending beyond the identification of potential inhibitors for SARS-CoV to contributing essential knowledge for the development of therapeutics against coronaviruses in general. This study serves as a crucial stepping stone for subsequent experimental validation and optimization of the identified compounds, propelling future drug discovery endeavours in the fight against coronaviral infections.

In silico approaches to investigate enzyme immobilization: a comprehensive systematic review.

Enzymes are popular catalysts with many applications, especially in industry. Biocatalyst usage on a large scale is facing some limitations, such as low operational stability, low recyclability, and high enzyme cost. Enzyme immobilization is a beneficial strategy to solve these problems. Bioinformatics tools can often correctly predict immobilization outcomes, resulting in a cost-effective experimental phase with the least time consumed. This study provides an overview of in silico methods predicting immobilization processes via a comprehensive systematic review of published articles till 11 December 2022. It also mentions the strengths and weaknesses of the processes and explains the computational analyses in each method that are required for immobilization assessment. In this regard, Web of Science and Scopus databases were screened to gain relevant publications. After screening the gathered documents (n = 3873), 60 articles were selected for the review. The selected papers have applied in silico procedures including only molecular dynamics (MD) simulations (n = 20), parallel tempering Monte Carlo (PTMC) and MD simulations (n = 3), MD and docking (n = 1), density functional theory (DFT) and MD (n = 1), only docking (n = 11), metal ion binding site prediction (MIB) server and docking (n = 2), docking and DFT (n = 1), docking and analysis of enzyme surfaces (n = 1), only DFT (n = 1), only MIB server (n = 2), analysis of an enzyme structure and surface (n = 12), rational design of immobilized derivatives (RDID) software (n = 3), and dissipative particle dynamics (DPD; n = 2). In most included studies (n = 51), enzyme immobilization was investigated experimentally in addition to in silico evaluation.

Design of some potent non-toxic Autoimmune disorder inhibitors based on 2D-QSAR, CoMFA, molecular docking, and molecular dynamics investigations

Current clinical research suggests that inhibitors of protein arginine deiminase 4 (PAD4), major histocompatibility complex (MHC) class II HLA-DQ-ALPHA chain, and thyrotropin receptor (or TSH receptor) which are of biological and therapeutic interest, may show potential in treating rheumatoid arthritis, type 1 diabetes, Graves' disease and other autoimmune disorder. In the present study, a comprehensive analysis was conducted on a collection of 32 compounds concerning their anti-rheumatoid arthritis activity as inhibitors of PAD4. This analysis represents the first instance in which these compounds were computationally examined, employing an in-silico approach that considered 2D-3D QSAR modeling, and molecular docking and was further validated through molecular dynamics and ADMET properties assessment. A credible 2D QSAR (Q_LOO^2 ​= ​0.6611 and R^2 ​= ​0.7535) model was constructed and verified using an external validation test set, Y-randomization, variance inflation factor (VIF), mean effect (MF), and William's plot applicability domain (AD). Ligand-based alignment was implemented in the 3D-QSAR examination. The outcomes demonstrated that CoMFA (uvepls) (Q2LOO ​= ​0.5877; R2 ​= ​0.9983) possess remarkable stability and foresight. The internal validation indicated that CoMFA (uvepls) MIFs display superior predictive capability compared to COMFA (ffdsel). Structural criteria determined by the contour maps of the model and molecular docking simulations were strategically employed to computationally develop 10 new, non-toxic autoimmune disease inhibitors with increased efficacy. Docking tests were done on the newly developed compounds to illustrate their binding mechanism and to identify critical interaction residues inside the active region of rheumatoid arthritis (PDB id: 3BLU). In addition, docking results of the selected designed compounds inside the active sites of type 1 diabetes receptor (6DFX), and Graves' disease receptor (4QT5) demonstrated their rheumatoid arthritis (PDB id: 3BLU) selectivity. A molecular dynamics simulation and binding free energy calculations using the MM/GBSA technique confirmed the stability of the proposed compound D4 inside the rheumatoid arthritis (3BLU) receptor active site. In summary, the results of our investigation might give considerable insight into the future design and development of new autoimmune disease inhibitors.

In silico-based Safety Assessment of the EPSPS CP4 single event expressed in genetically modified crops

In silico-based Safety Assessment of the EPSPS CP4 single event expressed in genetically modified crops

The potential of selenium to induce salt stress tolerance in Brassica rapa: Evaluation of biochemical, physiological and molecular phenomenon

The efficacy of selenium (Se) to ameliorate salinity stress in Brassica rapa was elucidated using an in-vitro and in-silico hybrid approach. Selenium was applied as a seed priming agent on the seeds grown in salt-spiked soils. The priming treatment facilitated all the growth indicators and physiological parameters, including chlorophyll synthesis, sugar contents, gas exchange attributes, etc. The advantageous effects of Se were accredited to the alleviation of salinity stress through reducing the malondialdehyde (MDA), proline, electrolyte leakage, and hydrogen peroxide (H2O2) level. Similarly, modulations in the expression of genes encoding antioxidative enzymes enhanced reactive oxygen species (ROS) scavenging capacity along with a decrease in sodium (Na+) ions, reciprocally enhancing potassium (K+) ions and resulting in a higher K+/Na+ ratio. Additionally, based on in silico studies, a significant difference in the surface overlap of the stress-responsive proteins, including DREB, SOS3, and STXBP1 of B. rapa, was observed, indicating the involvement of Se in the foiling interaction of NaCl with the enzymes. In the case of the CAT macromolecule, ligands NaCl exhibited a -1.9-fold lower docking score as compared to sodium selenite -4.9-fold, indicating a more overlapped interactive surface than sodium chloride. The binding energies analysis recorded -3.3 kcal/mol of binding energy for the best pose of sodium selenite in the case of DREB, while it was -3.5 and -3.5 kcal.mol-1 for SOS3 and 1 STXBP1, respectively. Ligand analysis revealed a frequent existence of hydrogen bonds besides the hydrophobic interactions, salt bridges, and pi–pi interactions representing high structural and functional stability. Results of the current study unveiled an in-vitro and in-silico hybrid bioinformatics system to unveil the biochemical, physiological and molecular phenomenon behind selenium-driven salt resilience in B. rapa.

In-silico analysis of probiotic attributes and safety assessment of probiotic strain Bacillus coagulans BCP92 for human application.

The whole genome sequence (WGS) of Bacillus coagulans BCP92 is reported along with its genomic analysis of probiotics and safety features. The identification of bacterial strain was carried out using the 16S rDNA sequencing method. Furthermore, gene-related probiotic features, safety assessment (by in vitro and in silico), and genome stability were also studied using the WGS analysis for the possible use of the bacterial strain as a probiotic. From the BLAST analysis, bacterial strain was identified as Bacillus (Heyndrickxia) coagulans. WGS analysis indicated that the genome consists of a 3 475 658bp and a GC-content of 46.35%. Genome mining of BCP92 revealed that the strain is consist of coding sequences for d-lactate dehydrogenase and l-lactate dehydrogenases, 36 genes involved in fermentation activities, 29 stress-responsive as well as many adhesions related genes. The genome, also possessing genes, is encoded for the synthesis of novel circular bacteriocin. Using an in-silico approach for the bacterial genome study, it was possible to determine that the Bacillus (Heyndrickxia) coagulans strain BCP92 contains genes that are encoded for the probiotic abilities and did not harbour genes that are risk associated, thus confirming the strain's safety and suitability as a probiotic to be used for human application.

Design and Discovery of Silmitasertib-based Drugs as a Potential Casein Kinase II Inhibitor for Cholangiocarcinoma through Hybrid In-silico Ligand-Based Virtual Screening with Molecular Docking Method

This research investigates the design and discovery of silmitasertib-based drugs as potential inhibitors of casein kinase II (CK2) for cholangiocarcinoma using hybrid in-silico methods. The study employs a two-fold approach, including ligand-based virtual screening and molecular docking, to assess the inhibitory potential of silmitasertib-based compounds. The ligand-based virtual screening involves the construction of a diverse compound library, followed by energy minimization and conformational analysis, revealing a set of compounds with varied similarities to silmitasertib. In parallel, the CK2 model undergoes meticulous selection, validation, and refinement through PDB REDO, showcasing significant improvements in model quality. Molecular docking results highlight promising candidates, with N-(2, 4, 6-trimethylphenyl) phenanthridin-6-amine (MCULE-1492185963-0) emerging as a lead compound with superior binding characteristics compared to silmitasertib. This hybrid in-silico approach demonstrates the potential of silmitasertib-based compounds as CK2 inhibitors for cholangiocarcinoma and identifies a lead compound for further experimental validation. The findings contribute valuable insights to the design of novel drugs for cholangiocarcinoma treatment, setting the stage for future experimental and clinical investigations

Unlocking the Potential of In-silico Approaches: Drug Development and Vaccine Design

Unmatched in its field, bioinformatics combines several academic fields such as statistics, computer science, mathematics, and biology to create state-of-the-art techniques for biological data retrieval, storage, and analysis that lead to a thorough understanding of the biological world. Countless options currently accessible in field of living sciences by the expansion of in-silico biology. Paradigm of life sciences has changed as a result of in-silico technologies, which offer researchers a valuable and affordable way to focus on in-silico techniques like homology modeling, epitope prediction, and molecular docking, which have impacted drug discovery and vaccine design. These techniques also provide previously unheard-of predictions and insights.

Insights into structural, vibrational, and chemical shift characteristics, solvents impact (polar and nonpolar) on electronic properties and reactive sites, ADMET predictions, and ligand-protein interactions for antiviral drugs safrole and isosafrole: An in-silico approach

The safrole and isosafrole are constitutional isomers that have been theoretically characterized utilizing Density Functional Theory (DFT) employed with the B3LYP/6-311++G(d,p) basis set to simulate geometrical parameters, vibrational aspects, electronic properties, and chemical shifts and explored the solvent effect on electronic spectra of safrole and isosafrole in different solvents, including polar solvents such as acetone, Dimethyl sulfoxide (DMSO), and methanol, nonpolar solvents such as chloroform, cyclohexane, and toluene, and the gas phase using Time Dependent-Density Functional Theory (TD-DFT) methods. In addition, Natural Orbital (NBO), Mulliken population analysis, and Molecular Electrostatic Potential (MEP) surface have been simulated to understand better the reactive sites in the molecular structures of safrole and isosafrole. The frontier molecular orbitals (FMO) energy gaps of safrole and isosafrole were simulated in polar and nonpolar solvents, confirming the safrole and isosafrole kinetic stability and chemical reactivity. The in-silico ADMET indices were utilized to ascertain the probable pharmacokinetic and biological features of safrole and isosafrole. A molecular docking investigation was also conducted to investigate the potential biological activity of safrole and isosafrole against the SARS-CoV-2 spike protein, providing more support for its antiviral efficacy.

GeNNius: an ultrafast drug-target interaction inference method based on graph neural networks.

Drug-target interaction (DTI) prediction is a relevant but challenging task in the drug repurposing field. In-silico approaches have drawn particular attention as they can reduce associated costs and time commitment of traditional methodologies. Yet, current state-of-the-art methods present several limitations: existing DTI prediction approaches are computationally expensive, thereby hindering the ability to use large networks and exploit available datasets and, the generalization to unseen datasets of DTI prediction methods remains unexplored, which could potentially improve the development processes of DTI inferring approaches in terms of accuracy and robustness. In this work, we introduce GeNNius (Graph Embedding Neural Network Interaction Uncovering System), a Graph Neural Network (GNN)-based method that outperforms state-of-the-art models in terms of both accuracy and time efficiency across a variety of datasets. We also demonstrated its prediction power to uncover new interactions by evaluating not previously known DTIs for each dataset. We further assessed the generalization capability of GeNNius by training and testing it on different datasets, showing that this framework can potentially improve the DTI prediction task by training on large datasets and testing on smaller ones. Finally, we investigated qualitatively the embeddings generated by GeNNius, revealing that the GNN encoder maintains biological information after the graph convolutions while diffusing this information through nodes, eventually distinguishing protein families in the node embedding space. GeNNius code is available at https://github.com/ubioinformat/GeNNius.