Protein Datasets Research Articles

Elliptic geometry-based kernel matrix for improved biological sequence classification

Protein sequence classification plays a pivotal role in bioinformatics as it enables the comprehension of protein functions and their involvement in diverse biological processes. While numerous machine learning models have been proposed to tackle this challenge, traditional approaches face limitations in capturing the intricate relationships and hierarchical structures inherent in genomic sequences. These limitations stem from operating within high-dimensional non-Euclidean spaces. To address this issue, we introduce the application of the elliptic geometry-based approach for protein sequence classification. First, we transform the problem in elliptic geometry and integrate it with the Gaussian kernel to map the problem into the Mercer kernel. The Gaussian-Elliptic approach allows for the implicit mapping of data into a higher-dimensional feature space, enabling the capture of complex nonlinear relationships. This feature becomes particularly advantageous when dealing with hierarchical or tree-like structures commonly encountered in biological sequences. Experimental results highlight the effectiveness of the proposed model in protein sequence classification, showcasing the advantages of utilizing elliptic geometry in bioinformatics analyses. It outperforms state-of-the-art methods by achieving 76% and 84% accuracies for DNA and Protein datasets, respectively. Furthermore, we provide theoretical justifications for the proposed model. This study contributes to the burgeoning field of geometric deep learning, offering insights into the potential applications of elliptic representations in the analysis of biological data.

Proteogenomic characterization of highly enriched viable leukemic blasts in acute myeloid leukemia: A SWOG report

AbstractIntroductionAcute myeloid leukemia (AML) remains one of the deadliest hematopoietic malignancies. A better understanding of the molecular biology governing AML may lead to improved risk stratification and facilitate the development of novel therapies. Proteins are responsible for much of the biology of cells. Several studies have examined the global proteome in bulk mononuclear cells (MNCs) from AML specimens, which are comprised a heterogenous population of cells at various stages of differentiation.MethodsGiven the potential impact of the nonleukemic cells on protein expression profiles, we applied an integrative proteogenomic approach utilizing next‐generation sequencing and mass spectrometry‐based proteomics to identify novel protein biomarkers in unsorted MNCs and viable leukemic blasts (VLBs) isolated from blood and bone marrow specimens obtained at the time of AML diagnosis.ResultsWe identified significant differences in protein expression between VLBs and MNCs. Subsequent studies (N = 27) focused on proteomic profiling of VLBs that identified novel candidate biomarkers associated with mutational genotypes and clinical outcome, some of which were recapitulated in an independent cohort of patients. Using mass spectrometry, we also detected mutated protein products, some of which were predicted via in silico analyses to be potential neoantigens amenable to adoptive immunotherapy. As previously described, analyses comparing transcript and protein expression showed an overall modest correlation between mRNA and protein dataset, but enriching for genes associated with mutations significantly improved the protein–RNA correlation.ConclusionTogether, the results provide insight into the biology of VLBs and demonstrate the gains derived from examining the proteome in addition to genome and transcriptome.

Proteomic Profiling of Endothelial Cell Secretomes After Exposure to Calciprotein Particles Reveals Downregulation of Basement Membrane Assembly and Increased Release of Soluble CD59

Calciprotein particles (CPPs) are essential circulating scavengers of excessive Ca2+ and PO43− ions, representing a vehicle that removes them from the human body and precludes extraskeletal calcification. Having been internalised by endothelial cells (ECs), CPPs induce their dysfunction, which is accompanied by a remarkable molecular reconfiguration, although little is known about this process’s extracellular signatures. Here, we applied ultra-high performance liquid chromatography-tandem mass spectrometry to perform a secretome-wide profiling of the cell culture supernatant from primary human coronary artery ECs (HCAECs) and internal thoracic artery ECs (HITAECs) treated with primary CPPs (CPP-P), secondary CPPs (CPP-S), magnesiprotein particles (MPPs), or Ca2+/Mg2+-free Dulbecco’s phosphate-buffered saline (DPBS) for 24 h. Incubation with CPP-P/CPP-S significantly altered the profiles of secreted proteins, delineating physiological and pathological endothelial secretomes. Neither pathway enrichment analysis nor the interrogation of protein–protein interactions detected extracellular matrix- and basement membrane-related molecular terms in the protein datasets from CPP-P/CPP-S-treated ECs. Both proteomic profiling and enzyme-linked immunosorbent assay identified an increased level of protectin (CD59) and reduced levels of osteonectin (SPARC), perlecan (HSPG2), and fibronectin (FN1) in the cell culture supernatant upon CPP-P/CPP-S treatment. Elevated soluble CD59 and decreased release of basement membrane components might be considered as potential signs of dysfunctional endothelium.

GraphPI: Efficient Protein Inference with Graph Neural Networks.

The integration of deep learning approaches in biomedical research has been transformative, enabling breakthroughs in various applications. Despite these strides, its application in protein inference is impeded by the scarcity of extensively labeled data sets, a challenge compounded by the high costs and complexities of accurate protein annotation. In this study, we introduce GraphPI, a novel framework that treats protein inference as a node classification problem. We treat proteins as interconnected nodes within a protein-peptide-PSM graph, utilizing a graph neural network-based architecture to elucidate their interrelations. To address label scarcity, we train the model on a set of unlabeled public protein data sets with pseudolabels derived from an existing protein inference algorithm, enhanced by self-training to iteratively refine labels based on confidence scores. Contrary to prevalent methodologies necessitating data set-specific training, our research illustrates that GraphPI, due to the well-normalized nature of Percolator features, exhibits universal applicability without data set-specific fine-tuning, a feature that not only mitigates the risk of overfitting but also enhances computational efficiency. Our empirical experiments reveal notable performance on various test data sets and deliver significantly reduced computation times compared to common protein inference algorithms.

Topology-Driven Discovery of Transmembrane Protein S-Palmitoylation.

Protein S-palmitoylation is a reversible lipophilic posttranslational modification regulating a diverse number of signaling pathways. Within transmembrane proteins (TMPs), S-palmitoylation is implicated in conditions from inflammatory disorders to respiratory viral infections. Many small-scale experiments have observed S-palmitoylation at juxtamembrane Cys residues. However, most large-scale S-palmitoyl discovery efforts rely on trypsin-based proteomics within which hydrophobic juxtamembrane regions are likely underrepresented. Machine learning- by virtue of its freedom from experimental constraints - is particularly well suited to address this discovery gap surrounding TMP S-palmitoylation. Utilizing a UniProt-derived feature set, a gradient boosted machine learning tool (TopoPalmTree) was constructed and applied to a holdout dataset of viral S-palmitoylated proteins. Upon application to the mouse TMP proteome, 1591 putative S-palmitoyl sites (i.e. not listed in SwissPalm or UniProt) were identified. Two lung-expressed S-palmitoyl candidates (synaptobrevin Vamp5 and water channel Aquaporin-5) were experimentally assessed. Finally, TopoPalmTree was used for rational design of an S-palmitoyl site on KDEL-Receptor 2. This readily interpretable model aligns the innumerable small-scale experiments observing juxtamembrane S-palmitoylation into a proteomic tool for TMP S-palmitoyl discovery and design, thus facilitating future investigations of this important modification.

Accurate and efficient protein embedding using multi-teacher distillation learning.

Protein embedding, which represents proteins as numerical vectors, is a crucial step in various learning-based protein annotation/classification problems, including gene ontology prediction, protein-protein interaction prediction, and protein structure prediction. However, existing protein embedding methods are often computationally expensive due to their large number of parameters, which can reach millions or even billions. The growing availability of large-scale protein datasets and the need for efficient analysis tools have created a pressing demand for efficient protein embedding methods. We propose a novel protein embedding approach based on multi-teacher distillation learning, which leverages the knowledge of multiple pre-trained protein embedding models to learn a compact and informative representation of proteins. Our method achieves comparable performance to state-of-the-art methods while significantly reducing computational costs and resource requirements. Specifically, our approach reduces computational time by ∼70% and maintains ±1.5% accuracy as the original large models. This makes our method well-suited for large-scale protein analysis and enables the bioinformatics community to perform protein embedding tasks more efficiently. The source code of MTDP is available via https://github.com/KennthShang/MTDP.

PB-GPT: An innovative GPT-based model for protein backbone generation

With advanced computational methods, it is now feasible to modify or design proteins for specific functions, a process with significant implications for disease treatment and other medical applications. Protein structures and functions are intrinsically linked to their backbones, making the design of these backbones a pivotal aspect of protein engineering. In this study, we focus on the task of unconditionally generating protein backbones. By means of codebook quantization and compression dictionaries, we convert protein backbone structures into a distinctive coded language and propose a GPT-based protein backbone generation model, PB-GPT. To validate the generalization performance of the model, we trained and evaluated the model on both public datasets and small protein datasets. The results demonstrate that our model has the capability to unconditionally generate elaborate, highly realistic protein backbones with structural patterns resembling those of natural proteins, thus showcasing the significant potential of large language models in protein structure design.

AlphaFold2-based prediction of the co-condensation propensity of proteins

The process of protein phase separation into liquid condensates has been implicated in the formation of membraneless organelles (MLOs), which selectively concentrate biomolecules to perform essential cellular functions. Although the importance of this process in health and disease is increasingly recognized, the experimental identification of proteins forming MLOs remains a complex challenge. In this study, we addressed this problem by harnessing the power of AlphaFold2 to perform computational predictions of the conformational properties of proteins from their amino acid sequences. We thus developed the CoDropleT (co-condensation into droplet transformer) method of predicting the propensity of co-condensation of protein pairs. The method was trained by combining experimental datasets of co-condensing proteins from the CD-CODE database with curated negative datasets of non-co-condensing proteins. To illustrate the performance of the method, we applied it to estimate the propensity of proteins to co-condense into MLOs. Our results suggest that CoDropleT could facilitate functional and therapeutic studies on protein condensation by predicting the composition of protein condensates.

Incorporating Tissue-Specific Gene Expression Data to Improve Chemical-Disease Inference of in Silico Toxicogenomics Methods.

In silico toxicogenomics methods are resource- and time-efficient approaches for inferring chemical-protein-disease associations with potential mechanism information for exploring toxicological effects. However, current in silico toxicogenomics systems make inferences based on only chemical-protein interactions without considering tissue-specific gene/protein expressions. As a result, inferred diseases could be overpredicted with false positives. In this work, six tissue-specific expression datasets of genes and proteins were collected from the Expression Atlas. Genes were then categorized into high, medium, and low expression levels in a tissue- and dataset-specific manner. Subsequently, the tissue-specific expression datasets were incorporated into the chemical-protein-disease inference process of our ChemDIS system by filtering out relatively low-expressed genes. By incorporating tissue-specific gene/protein expression data, the enrichment rate for chemical-disease inference was largely improved with up to 62.26% improvement. A case study of melamine showed the ability of the proposed method to identify more specific disease terms that are consistent with the literature. A user-friendly user interface was implemented in the ChemDIS system. The methodology is expected to be useful for chemical-disease inference and can be implemented for other in silico toxicogenomics tools.

Correction of preferred orientation-induced distortion in cryo-electron microscopy maps.

Reconstruction maps of cryo-electron microscopy (cryo-EM) exhibit distortion when the cryo-EM dataset is incomplete, usually caused by unevenly distributed orientations. Prior efforts had been attempted to address this preferred orientation problem using tilt-collection strategy and modifications to grids or to air-water interfaces. However, these approaches often require time-consuming experiments, and the effect was always protein dependent. Here, we developed a procedure containing removing misaligned particles and an iterative reconstruction method based on signal-to-noise ratio of Fourier component to correct this distortion by recovering missing data using a purely computational algorithm. This procedure called signal-to-noise ratio iterative reconstruction method (SIRM) was applied on incomplete datasets of various proteins to fix distortion in cryo-EM maps and to a more isotropic resolution. In addition, SIRM provides a better reference map for further reconstruction refinements, resulting in an improved alignment, which ultimately improves map quality and benefits model building.

Secretome analysis of oligodendrocytes and precursors reveals their roles as contributors to the extracellular matrix and potential regulators of inflammation.

Oligodendrocytes form myelin that ensheaths axons and accelerates the speed of action potential propagation. Oligodendrocyte progenitor cells (OPCs) proliferate and replenish oligodendrocytes. While the myelin-forming role of oligodendrocytes and OPCs is well-established, potential additional roles of these cells are yet to be fully explored. Here, we analyzed the secreted proteome of oligodendrocytes and OPCs in vitro to determine whether these cell types are major sources of secreted proteins in the central nervous system (CNS). Interestingly, we found that both oligodendrocytes and OPCs secret various extracellular matrix proteins. Considering the critical role of neuroinflammation in neurological disorders, we evaluated the responses and potential contributions of oligodendrocytes and OPCs to this process. By characterizing the secreted proteomes of these cells after pro-inflammatory cytokine treatment, we discovered the secretion of immunoregulators such as C2 and B2m. This finding sheds new light on the hitherto underappreciated role of oligodendrocytes and OPCs in actively modulating neuroinflammation. Our study provides a comprehensive and unbiased proteomic dataset of proteins secreted by oligodendrocyte and OPC under both physiological and inflammatory conditions. It revealed the potential of these cells to secrete matrix and signaling molecules, highlighting their multifaceted function beyond their conventional myelin-forming roles.

ProteinFlow: An advanced framework for feature engineering in protein data analysis.

In the burgeoning field of proteins, the effective analysis of intricate protein data remains a formidable challenge, necessitating advanced computational tools for data processing, feature extraction, and interpretation. This study introduces ProteinFlow, an innovative framework designed to revolutionize feature engineering in protein data analysis. ProteinFlow stands out by offering enhanced efficiency in data collection and preprocessing, along with advanced capabilities in feature extraction, directly addressing the complexities inherent in multidimensional protein data sets. Through a comparative analysis, ProteinFlow demonstrated a significant improvement over traditional methods, notably reducing data preprocessing time and expanding the scope of biologically significant features identified. The framework's parallel data processing strategy and advanced algorithms ensure not only rapid data handling but also the extraction of comprehensive, meaningful insights from protein sequences, structures, and interactions. Furthermore, ProteinFlow exhibits remarkable scalability, adeptly managing large-scale data sets without compromising performance, a crucial attribute in the era of big data.

Large-scale identification of calcium oxalate stone inhibitory proteins in normal human urine

Recent evidence has shown that proteins in normal human urine can inhibit calcium oxalate (CaOx) kidney stone formation. Herein, we performed fast protein liquid chromatography (FPLC) to fractionate normal human urinary proteins using anion-exchange (DEAE) and size-exclusion (Superdex 200) materials. FPLC fractions (F1-F15) were examined by CaOx crystallization, growth, aggregation and crystal-cell adhesion assays. The fractions with potent inhibitory activities against CaOx crystals were then subjected to mass spectrometric protein identification. The data revealed that 13 of 15 fractions showed inhibitory activities in at least one crystal assay. Integrating CaOx inhibitory scores demonstrated that F6, F7 and F8 had the most potent inhibitory activities. NanoLC-ESI-Qq-TOF MS/MS identified 105, 93 and 53 proteins in F6, F7 and F8, respectively. Among them, 60 were found in at least two fractions and/or listed among known inhibitors with solid experimental evidence in the StoneMod database (https://www.stonemod.org). Interestingly, 10 of these 60 potential inhibitors have been reported with lower urinary levels in CaOx stone formers compared with healthy (non-stone) individuals, strengthening their roles as potent CaOx stone inhibitors. Our study provides the largest dataset of potential CaOx stone inhibitory proteins that will be useful for further elucidations of stone-forming mechanisms and ultimately for therapeutic/preventive applications.

Optimal phylogenetic reconstruction of insertion and deletion events.

Insertions and deletions (indels) influence the genetic code in fundamentally distinct ways from substitutions, significantly impacting gene product structure and function. Despite their influence, the evolutionary history of indels is often neglected in phylogenetic tree inference and ancestral sequence reconstruction, hindering efforts to comprehend biological diversity determinants and engineer variants for medical and industrial applications. We frame determining the optimal history of indel events as a single Mixed-Integer Programming (MIP) problem, across all branch points in a phylogenetic tree adhering to topological constraints, and all sites implied by a given set of aligned, extant sequences. By disentangling the impact on ancestral sequences at each branch point, this approach identifies the minimal indel events that jointly explain the diversity in sequences mapped to the tips of that tree. MIP can recover alternate optimal indel histories, if available. We evaluated MIP for indel inference on a dataset comprising 15 real phylogenetic trees associated with protein families ranging from 165 to 2000 extant sequences, and on 60 synthetic trees at comparable scales of data and reflecting realistic rates of mutation. Across relevant metrics, MIP outperformed alternative parsimony-based approaches and reported the fewest indel events, on par or below their occurrence in synthetic datasets. MIP offers a rational justification for indel patterns in extant sequences; importantly, it uniquely identifies global optima on complex protein data sets without making unrealistic assumptions of independence or evolutionary underpinnings, promising a deeper understanding of molecular evolution and aiding novel protein design. The implementation is available via GitHub at https://github.com/santule/indelmip.

Pairing interacting protein sequences using masked language modeling

Predicting which proteins interact together from amino acid sequences is an important task. We develop a method to pair interacting protein sequences which leverages the power of protein language models trained on multiple sequence alignments (MSAs), such as MSA Transformer and the EvoFormer module of AlphaFold. We formulate the problem of pairing interacting partners among the paralogs of two protein families in a differentiable way. We introduce a method called Differentiable Pairing using Alignment-based Language Models (DiffPALM) that solves it by exploiting the ability of MSA Transformer to fill in masked amino acids in multiple sequence alignments using the surrounding context. MSA Transformer encodes coevolution between functionally or structurally coupled amino acids within protein chains. It also captures inter-chain coevolution, despite being trained on single-chain data. Relying on MSA Transformer without fine-tuning, DiffPALM outperforms existing coevolution-based pairing methods on difficult benchmarks of shallow multiple sequence alignments extracted from ubiquitous prokaryotic protein datasets. It also outperforms an alternative method based on a state-of-the-art protein language model trained on single sequences. Paired alignments of interacting protein sequences are a crucial ingredient of supervised deep learning methods to predict the three-dimensional structure of protein complexes. Starting from sequences paired by DiffPALM substantially improves the structure prediction of some eukaryotic protein complexes by AlphaFold-Multimer. It also achieves competitive performance with using orthology-based pairing.

A convenient model of serum-induced reactivity of human astrocytes to investigate astrocyte-derived extracellular vesicles.

Extracellular vesicles (EVs) are secreted by all cells in the CNS, including neurons and astrocytes. EVs are lipid membrane enclosed particles loaded with various bioactive cargoes reflecting the dynamic activities of cells of origin. In contrast to neurons, the specific role of EVs released by astrocytes is less well understood, partly due to the difficulty in maintaining primary astrocyte cultures in a quiescent state. The aim of this study was to establish a human serum-free astrocyte culture system that maintains primary astrocytes in a quiescent state to study the morphology, function, and protein cargoes of astrocyte-derived EVs. Serum-free medium with G5 supplement and serum-supplemented medium with 2% FBS were compared for the culture of commercially available human primary fetal astrocytes. Serum-free astrocytes displayed morphologies similar to in vivo astrocytes, and surprisingly, higher levels of astrocyte markers compared to astrocytes chronically cultured in FBS. In contrast, astrocyte and inflammatory markers in serum-free astrocytes were upregulated 24 h after either acute 2% FBS or cytokine exposure, confirming their capacity to become reactive. Importantly, this suggests that distinct signaling pathways are involved in acute and chronic astrocyte reactivity. Despite having a similar morphology, chronically serum-cultured astrocyte-derived EVs (ADEVs) were smaller in size compared to serum-free ADEVs and could reactivate serum-free astrocytes. Proteomic analysis identified distinct protein datasets for both types of ADEVs with enrichment of complement and coagulation cascades for chronically serum-cultured astrocyte-derived EVs, offering insights into their roles in the CNS. Collectively, these results suggest that human primary astrocytes cultured in serum-free medium bear similarities with in vivo quiescent astrocytes and the addition of serum induces multiple morphological and transcriptional changes that are specific to human reactive astrocytes and their ADEVs. Thus, more emphasis should be made on using multiple structural, molecular, and functional parameters when evaluating ADEVs as biomarkers of astrocyte health.

ProMENDA: an updated resource for proteomic and metabolomic characterization in depression

Depression is a prevalent mental disorder with a complex biological mechanism. Following the rapid development of systems biology technology, a growing number of studies have applied proteomics and metabolomics to explore the molecular profiles of depression. However, a standardized resource facilitating the identification and annotation of the available knowledge from these scattered studies associated with depression is currently lacking. This study presents ProMENDA, an upgraded resource that provides a platform for manual annotation of candidate proteins and metabolites linked to depression. Following the establishment of the protein dataset and the update of the metabolite dataset, the ProMENDA database was developed as a major extension of its initial release. A multi-faceted annotation scheme was employed to provide comprehensive knowledge of the molecules and studies. A new web interface was also developed to improve the user experience. The ProMENDA database now contains 43,366 molecular entries, comprising 20,847 protein entries and 22,519 metabolite entries, which were manually curated from 1370 human, rat, mouse, and non-human primate studies. This represents a significant increase (more than 7-fold) in molecular entries compared to the initial release. To demonstrate the usage of ProMENDA, a case study identifying consistently reported proteins and metabolites in the brains of animal models of depression was presented. Overall, ProMENDA is a comprehensive resource that offers a panoramic view of proteomic and metabolomic knowledge in depression. ProMENDA is freely available at https://menda.cqmu.edu.cn.

Application of artificial intelligence and machine learning techniques to the analysis of dynamic protein sequences.

We apply methods of Artificial Intelligence and Machine Learning to protein dynamic bioinformatics. We rewrite the sequences of a large protein data set, containing both folded and intrinsically disordered molecules, using a representation developed previously, which encodes the intrinsic dynamic properties of the naturally occurring amino acids. We Fourier analyze the resulting sequences. It is demonstrated that classification models built using several different supervised learning methods are able to successfully distinguish folded from intrinsically disordered proteins from sequence alone. It is further shown that the most important sequence property for this discrimination is the sequence mobility, which is the sequence averaged value of the residue-specific average alpha carbon B factor. This is in agreement with previous work, in which we have demonstrated the central role played by the sequence mobility in protein dynamic bioinformatics and biophysics. This finding opens a path to the application of dynamic bioinformatics, in combination with machine learning algorithms, to a range of significant biomedical problems.

Long-term efficacy and safety of cow’s milk anaphylaxis specific immunotherapy: Allergen unresponsiveness via the Tolerance Induction Program

Tolerance Induction Program (TIP) immunotherapy applies machine learning contextualized on immunologic and food protein data sets. TIP has established efficacy toward peanut allergy. This form of treatment demonstrates equal efficacy toward cow's milk anaphylaxis. TIP maintains remission outcomes defined as a minimum of 7 days of allergen unresponsiveness to high-dose protein exposures. Furthermore, remission patients openly consume unrestricted amounts of dairy protein. We sought to assess the rate of decline in specific IgE specific whole and component-resolved diagnostics following 1 year of TIP milk immunotherapy. The study comprised 214 cow milk anaphylactic children who underwent TIP at the Translational Pulmonary & Immunology Research Center/Food Allergy Institute. Postintervention changes in cow milk specific IgE, component-resolved diagnostics, and specific IgG4 were assessed. After 1 year of 10-g dairy protein weekly sustained unresponsiveness, eosinophil count decreased from 558.38 to 409.26 cells/μL, the mean cow milk IgE decreased from 16.91 to 9.10 kU/L, the mean boiled cow milk IgE decreased from 12.89 to 6.03 kU/L, the mean Bos D4 decreased from 7.38 to 3.52 kU/L, the mean Bos D5 decreased from 6.79 to 3.16 kU/L, and the mean Bos D8 decreased from 13.55 to 6.62 kU/L. Adverse events were rare. TIP cow milk immunotherapy significantly reduced cow milk specific IgE and component-resolved diagnostics while increasing specific IgG4 in cow milk anaphylactic children. TIP demonstrates safety and clinical efficacy in cow milk anaphylaxis treatment.

VISH-Pred: an ensemble of fine-tuned ESM models for protein toxicity prediction.

Peptide- and protein-based therapeutics are becoming a promising treatment regimen for myriad diseases. Toxicity of proteins is the primary hurdle for protein-based therapies. Thus, there is an urgent need for accurate in silico methods for determining toxic proteins to filter the pool of potential candidates. At the same time, it is imperative to precisely identify non-toxic proteins to expand the possibilities for protein-based biologics. To address this challenge, we proposed an ensemble framework, called VISH-Pred, comprising models built by fine-tuning ESM2 transformer models on a large, experimentally validated, curated dataset of protein and peptide toxicities. The primary steps in the VISH-Pred framework are to efficiently estimate protein toxicities taking just the protein sequence as input, employing an under sampling technique to handle the humongous class-imbalance in the data and learning representations from fine-tuned ESM2 protein language models which are then fed to machine learning techniques such as Lightgbm and XGBoost. The VISH-Pred framework is able to correctly identify both peptides/proteins with potential toxicity and non-toxic proteins, achieving a Matthews correlation coefficient of 0.737, 0.716 and 0.322 and F1-score of 0.759, 0.696 and 0.713 on three non-redundant blind tests, respectively, outperforming other methods by over $10\%$ on these quality metrics. Moreover, VISH-Pred achieved the best accuracy and area under receiver operating curve scores on these independent test sets, highlighting the robustness and generalization capability of the framework. By making VISH-Pred available as an easy-to-use web server, we expect it to serve as a valuable asset for future endeavors aimed at discerning the toxicity of peptides and enabling efficient protein-based therapeutics.