Human Protein-protein Interaction Research Articles

WeSA: a web server for improving analysis of affinity proteomics data.

Protein-protein interaction experiments still yield many false positive interactions. The socioaffinity metric can distinguish true protein-protein interactions from noise based on available data. Here, we present WeSA (Weighted SocioAffinity), which considers large datasets of interaction proteomics data (IntAct, BioGRID, the BioPlex) to score human protein interactions and, in a statistically robust way, flag those (even from a single experiment) that are likely to be false positives. ROC analysis (using CORUM-PDB positives and Negatome negatives) shows that WeSA improves over other measures of interaction confidence. WeSA shows consistently good results over all datasets (up to: AUC=0.93 and at best threshold: TPR=0.84, FPR=0.11, Precision=0.98). WeSA is freely available without login (wesa.russelllab.org). Users can submit their own data or look for organized information on human protein interactions using the web server. Users can either retrieve available information for a list of proteins of interest or calculate scores for new experiments. The server outputs either pre-computed or updated WeSA scores for the input enriched with information from databases. The summary is presented as a table and a network-based visualization allowing the user to remove those nodes/edges that the method considers spurious.

Profiling Protein-Protein Interactions in the Human Brain by Refined Cofractionation Mass Spectrometry.

Proteins usually execute their biological functions through interactions with other proteins and by forming macromolecular complexes, but global profiling of protein complexes directly from human tissue samples has been limited. In this study, we utilized cofractionation mass spectrometry (CF-MS) to map protein complexes within the postmortem human brain with experimental replicates. First, we used concatenated anion and cation Ion Exchange Chromatography (IEX) to separate native protein complexes in 192 fractions and then proceeded with Data-Independent Acquisition (DIA) mass spectrometry to analyze the proteins in each fraction, quantifying a total of 4,804 proteins with 3,260 overlapping in both replicates. We improved the DIA's quantitative accuracy by implementing a constant amount of bovine serum albumin (BSA) in each fraction as an internal standard. Next, advanced computational pipelines, which integrate both a database-based complex analysis and an unbiased protein-protein interaction (PPI) search, were applied to identify protein complexes and construct protein-protein interaction networks in the human brain. Our study led to the identification of 486 protein complexes and 10054 binary protein-protein interactions, which represents the first global profiling of human brain PPIs using CF-MS. Overall, this study offers a resource and tool for a wide range of human brain research, including the identification of disease-specific protein complexes in the future.

Identification of Protein-Protein Interaction Associated Functions Based on Gene Ontology.

Protein-protein interactions (PPIs) involve the physical or functional contact between two or more proteins. Generally, proteins that can interact with each other always have special relationships. Some previous studies have reported that gene ontology (GO) terms are related to the determination of PPIs, suggesting the special patterns on the GO terms of proteins in PPIs. In this study, we explored the special GO term patterns on human PPIs, trying to uncover the underlying functional mechanism of PPIs. The experimental validated human PPIs were retrieved from STRING database, which were termed as positive samples. Additionally, we randomly paired proteins occurring in positive samples, yielding lots of negative samples. A simple calculation was conducted to count the number of positive samples for each GO term pair, where proteins in samples were annotated by GO terms in the pair individually. The similar number for negative samples was also counted and further adjusted due to the great gap between the numbers of positive and negative samples. The difference of the above two numbers and the relative ratio compared with the number on positive samples were calculated. This ratio provided a precise evaluation of the occurrence of GO term pairs for positive samples and negative samples, indicating the latent GO term patterns for PPIs. Our analysis unveiled several nuclear biological processes, including gene transcription, cell proliferation, and nutrient metabolism, as key biological functions. Interactions between major proliferative or metabolic GO terms consistently correspond with significantly reported PPIs in recent literature.

Purification and characterisation of the platelet-activating GPVI/FcRγ complex in SMALPs

The collagen/fibrin(ogen) receptor, glycoprotein VI (GPVI), is a platelet activating receptor and a promising anti-thrombotic drug target. However, while agonist-induced GPVI clustering on platelet membranes has been shown to be essential for its activation, it is unknown if GPVI dimerisation represents a unique conformation for ligand binding. Current GPVI structures all contain only the two immunoglobulin superfamily (IgSF) domains in the GPVI extracellular region, so lacking the mucin-like stalk, transmembrane, cytoplasmic tail of GPVI and its associated Fc receptor γ (FcRγ) homodimer signalling chain, and provide contradictory insights into the mechanisms of GPVI dimerisation. Here, we utilised styrene maleic-acid lipid particles (SMALPs) to extract GPVI in complex with its two associated FcRγ chains from transfected HEK-293T cells, together with the adjacent lipid bilayer, then purified and characterised the GPVI/FcRγ-containing SMALPs, to enable structural insights into the full-length GPVI/FcRγ complex. Using size exclusion chromatography followed by a native polyacrylamide gel electrophoresis (PAGE) method, SMA-PAGE, we revealed multiple sizes of the purified GPVI/FcRγ SMALPs, suggesting the potential existence of GPVI oligomers. Importantly, GPVI/FcRγ SMALPs were functional as they could bind collagen. Mono-dispersed GPVI/FcRγ SMALPs could be observed under negative stain electron microscopy. These results pave the way for the future investigation of GPVI stoichiometry and structure, while also validating SMALPs as a promising tool for the investigation of human membrane protein interactions, stoichiometry and structure.

SynerGNet: A Graph Neural Network Model to Predict Anticancer Drug Synergy.

Drug combination therapy shows promise in cancer treatment by addressing drug resistance, reducing toxicity, and enhancing therapeutic efficacy. However, the intricate and dynamic nature of biological systems makes identifying potential synergistic drugs a costly and time-consuming endeavor. To facilitate the development of combination therapy, techniques employing artificial intelligence have emerged as a transformative solution, providing a sophisticated avenue for advancing existing therapeutic approaches. In this study, we developed SynerGNet, a graph neural network model designed to accurately predict the synergistic effect of drug pairs against cancer cell lines. SynerGNet utilizes cancer-specific featured graphs created by integrating heterogeneous biological features into the human protein-protein interaction network, followed by a reduction process to enhance topological diversity. Leveraging synergy data provided by AZ-DREAM Challenges, the model yields a balanced accuracy of 0.68, significantly outperforming traditional machine learning. Encouragingly, augmenting the training data with carefully constructed synthetic instances improved the balanced accuracy of SynerGNet to 0.73. Finally, the results of an independent validation conducted against DrugCombDB demonstrated that it exhibits a strong performance when applied to unseen data. SynerGNet shows a great potential in detecting drug synergy, positioning itself as a valuable tool that could contribute to the advancement of combination therapy for cancer treatment.

P969 Unveiling the molecular mechanisms of the combination of vedolizumab with JAK inhibitors in Crohn’s Disease through a systems biology and artificial intelligence-based approach

Abstract Background Clinical evidence supports the combination therapy of vedolizumab (a gut-selective, anti-lymphocyte trafficking drug inhibiting α4β7-integrin) with JAK 1, 2, 3 inhibitors (JAKi) for Inflammatory Bowel Disease. However, the mechanistic explanation for the improved outcomes in Ulcerative Colitis patients treated with vedolizumab plus JAKi remains unknown and even more in Crohn's Disease (CD). In this study we used an unbiased in silico systems biology and artificial intelligence approach to gain a holistic view of the mechanisms that may underpin the potential benefits of the mentioned combination therapy in the treatment of CD. Methods The drugs under study -vedolizumab and JAKi- and CD were characterized at protein level by compiling data from extensive literature search. Through it, human protein interaction networks focused on CD protein effectors were generated and used to construct mathematical models using Therapeutic Performance Mapping System (TPMS) technology (Anaxomics Biotech, Barcelona, Spain). Sampling-based methods were employed to assess the impact on CD of the combined therapy vedolizumab plus JAKi and to describe underlying molecular mechanisms. Results Characterization of CD allowed identifying 4 pathological processes (referred to as motives-M) underlying the manifestation of the disease: intestinal barrier disruption (M1), increased innate immune response (M2), chronic inflammation and Th1/Th17 adaptive immune response (M3) and tissue remodeling (M4) and the molecular effectors or proteins involved in each motive. The combination of vedolizumab and JAKi induced a high degree of reversion of the effectors altered in CD (54.76%), higher than the percentage reverted by the individual drugs. Combined vedolizumab plus JAKi therapy mainly modulated M3, where both drugs mostly converge although by distinct mechanisms, thus providing additional individual benefits. The combination therapy reverted a great percentage of M3 effectors, for example, CCR9 and FASLG which were increased in CD, were reverted (or downregulated). In addition to M3 modulation, the combination therapy also modulated M1 and M4. In M1, vedolizumab is the one that would provide more complementary mechanisms apart from those that already converge in both drugs. Regarding M4, it seems that main contribution comes from complementarity pathways of the different mechanisms of action of each individual drug. Conclusion The constructed in silico models reveal that the combined vedolizumab plus JAKi therapy could modulate a high array of effectors altered in CD, more than any of the therapies alone, by enhancing the effects of the individual drugs, and provide a mechanistic rationale for employing the combination of these drugs as potential therapy for CD.

Pitfalls of machine learning models for protein-protein interaction networks.

Protein-protein interactions (PPIs) are essential to understanding biological pathways as well as their roles in development and disease. Computational tools, based on classic machine learning, have been successful at predicting PPIs in silico, but the lack of consistent and reliable frameworks for this task has led to network models that are difficult to compare and discrepancies between algorithms that remain unexplained. To better understand the underlying inference mechanisms that underpin these models, we designed an open-source framework for benchmarking that accounts for a range of biological and statistical pitfalls while facilitating reproducibility. We use it to shed light on the impact of network topology and how different algorithms deal with highly connected proteins. By studying functional genomics-based and sequence-based models on human PPIs, we show their complementarity as the former performs best on lone proteins while the latter specializes in interactions involving hubs. We also show that algorithm design has little impact on performance with functional genomic data. We replicate our results between both human and S. cerevisiae data and demonstrate that models using functional genomics are better suited to PPI prediction across species. With rapidly increasing amounts of sequence and functional genomics data, our study provides a principled foundation for future construction, comparison, and application of PPI networks. The code and data are available on GitHub: https://github.com/Llannelongue/B4PPI.

Identification of oral therapeutics using an AI platform against the virus responsible for COVID-19, SARS-CoV-2.

Purpose: This study introduces a sophisticated computational pipeline, eVir, designed for the discovery of antiviral drugs based on their interactions within the human protein network. There is a pressing need for cost-effective therapeutics for infectious diseases (e.g., COVID-19), particularly in resource-limited countries. Therefore, our team devised an Artificial Intelligence (AI) system to explore repurposing opportunities for currently used oral therapies. The eVir system operates by identifying pharmaceutical compounds that mirror the effects of antiviral peptides (AVPs)-fragments of human proteins known to interfere with fundamental phases of the viral life cycle: entry, fusion, and replication. eVir extrapolates the probable antiviral efficacy of a given compound by analyzing its established and predicted impacts on the human protein-protein interaction network. This innovative approach provides a promising platform for drug repurposing against SARS-CoV-2 or any virus for which peptide data is available. Methods: The eVir AI software pipeline processes drug-protein and protein-protein interaction networks generated from open-source datasets. eVir uses Node2Vec, a graph embedding technique, to understand the nuanced connections among drugs and proteins. The embeddings are input a Siamese Network (SNet) and MLPs, each tailored for the specific mechanisms of entry, fusion, and replication, to evaluate the similarity between drugs and AVPs. Scores generated from the SNet and MLPs undergo a Platt probability calibration and are combined into a unified score that gauges the potential antiviral efficacy of a drug. This integrated approach seeks to boost drug identification confidence, offering a potential solution for detecting therapeutic candidates with pronounced antiviral potency. Once identified a number of compounds were tested for efficacy and toxicity in lung carcinoma cells (Calu-3) infected with SARS-CoV-2. A lead compound was further identified to determine its efficacy and toxicity in K18-hACE2 mice infected with SARS-CoV-2. Computational Predictions: The SNet confidently differentiated between similar and dissimilar drug pairs with an accuracy of 97.28% and AUC of 99.47%. Key compounds identified through these networks included Zinc, Mebendazole, Levomenol, Gefitinib, Niclosamide, and Imatinib. Notably, Mebendazole and Zinc showcased the highest similarity scores, while Imatinib, Levemenol, and Gefitinib also ranked within the top 20, suggesting their significant pharmacological potentials. Further examination of protein binding analysis using explainable AI focused on reverse engineering the causality of the networks. Protein interaction scores for Mebendazole and Imatinib revealed their effects on notable proteins such as CDPK1, VEGF2, ABL1, and several tyrosine protein kinases. Laboratory Studies: This study determined that Mebendazole, Gefitinib, Topotecan and to some extent Carfilzomib showed conventional drug-response curves, with IC50 values near or below that of Remdesivir with excellent confidence all above R2>0.91, and no cytotoxicity at the IC50 concentration in Calu-3 cells. Cyclosporine A showed antiviral activity, but also unconventional drug-response curves and low R2 which are explained by the non-dose dependent toxicity of the compound. Additionally, Niclosamide demonstrated a conventional drug-response curve with high confidence; however, its inherent cytotoxicity may be a confounding element that misrepresents true antiviral efficacy, by reflecting cellular damage rather than a genuine antiviral action. Remdesivir was used as a control compound and was evaluated in parallel with the submitted test article and had conventional drug-response curves validating the overall results of the assay. Mebendazole was identified from the cell studies to have efficacy at non-toxic concentrations and were further evaluated in mice infected with SARS-CoV-2. Mebendazole administered to K18-hACE2 mice infected with SARS-CoV-2, resulted in a 44.2% reduction in lung viral load compared to non-treated placebo control respectively. There were no significant differences in body weight and all clinical chemistry determinations evaluated (i.e., kidney and liver enzymes) between the different treatment groups. Conclusion: This research underscores the potential of repurposing existing compounds for treating COVID-19. Our preliminary findings underscore the therapeutic promise of several compounds, notably Mebendazole, in both in vitro and in vivo settings against SARS-CoV-2. Several of the drugs explored, especially Mebendazole, are off-label medication; their cost-effectiveness position them as economical therapies against SARS-CoV-2.

Minor intron-containing genes as an ancient backbone for viral infection?

Minor intron-containing genes (MIGs) account for <2% of all human protein-coding genes and are uniquely dependent on the minor spliceosome for proper excision. Despite their low numbers, we surprisingly found a significant enrichment of MIG-encoded proteins (MIG-Ps) in protein-protein interactomes and host factors of positive-sense RNA viruses, including SARS-CoV-1, SARS-CoV-2, MERS coronavirus, and Zika virus. Similarly, we observed a significant enrichment of MIG-Ps in the interactomes and sets of host factors of negative-sense RNA viruses such as Ebola virus, influenza A virus, and the retrovirus HIV-1. We also found an enrichment of MIG-Ps in double-stranded DNA viruses such as Epstein-Barr virus, human papillomavirus, and herpes simplex viruses. In general, MIG-Ps were highly connected and placed in central positions in a network of human-host protein interactions. Moreover, MIG-Ps that interact with viral proteins were enriched with essential genes. We also provide evidence that viral proteins interact with ancestral MIGs that date back to unicellular organisms and are mainly involved in basic cellular functions such as cell cycle, cell division, and signal transduction. Our results suggest that MIG-Ps form a stable, evolutionarily conserved backbone that viruses putatively tap to invade and propagate in human host cells.

Superhydrophobic polyether sulfone (PES) dialysis membrane with enhanced hemocompatibility and reduced human serum protein interactions: Ex vivo, in situ synchrotron imaging, experimental, and computational studies

This study aimed to enhance the hemocompatibility of polyethersulfone (PES) membranes by increasing their hydrophobicity. Two methods were used: chemical modification with a hydrophobic substance and creating a micropatterned structure using modified silica nanoparticles (SiO2 NP) in a mixed matrix membrane (MMM). Fibrinogen adsorption was assessed using synchrotron radiation micro-computed tomography (SR-µCT), and surface roughness was analyzed with atomic force microscopy (AFM). Hemocompatibility was evaluated by studying interactions with patients' serum and measuring the release of inflammatory biomarkers (C5a, IL-1α, IL-1β, IL-6, vWF, C5b-9 and PF-4). Differential Scanning Calorimeter (DSC) analysis determined the stability of the membrane hydration layer, and water contact angle (WCA) measurements assessed hydrophobic properties. Fourier-transform infrared spectroscopy (FTIR) and surface charge analysis examined surface stability. Molecular docking analysis showed reduced interaction between hydrophobized PES membranes and fibrinogen and albumin in human serum. SR-µCT analysis revealed 20% less fibrinogen adsorption on hydrophobized PES membrane surfaces. MMM PES membranes exhibited minimal fibrinogen adsorption due to the micropatterned structure. Both hydrophobized PES membranes demonstrated reduced vWF biomarker release (25–28%) compared to unmodified PES, with MMM PES membranes performing best. However, both hydrophobized PES membranes led to a significant increase (150%) in IL-6 upon blood contact, attributed to the reduced hydration layer. MMM PES membranes contained the least strongly bound water (0.2%) and exhibited the highest hydrophobicity (98°). This study highlights the potential of hydrophobized PES membranes and MMM structures for enhancing hemocompatibility, with opportunities for optimizing surface modifications to reduce inflammation responses while maintaining desirable properties.

Interaction of Human Serum and Plasma Proteins with Polyelectrolyte Microcapsules of Different Structures

Interaction of Human Serum and Plasma Proteins with Polyelectrolyte Microcapsules of Different Structures

MyD88 in myeloid cells promotes angiotensin II-induced vascular inflammation and is associated with all-cause mortality and incident heart failure in humans with arterial hypertension

Abstract Background Angiotensin II (AngII) causes hypertension and promotes vascular accumulation of inflammatory cells. We aimed to determine how myeloid factor of differentiation 88 (MyD88) contributes to vascular dysfunction and arterial hypertension in mice and humans. Methods Male C57BL/6 as well as MyD88-/-, LysMCre/wtMyD88LSL/LSL, LysMCre/wt, TLR2-/-, TLR4-/-, TLR7-/- and TLR9-/- were investigated in the AngII-model of hypertension (1 mg/kg/d for 7days). Biodata analyses were performed in the Gutenberg-Health Study (GHS), a population based cohort. In addition, we performed computational analyses of known hypertension-related genes from genome-wide association studies (GWAS) to investigate whether they preferentially interact with MyD88 in the human protein network. Results MyD88 deficiency attenuated AngII-induced blood pressure increase and endothelial dysfunction in conductance and resistance vessels. Vascular mRNA expression levels of vcam-1, nos2, nox2 were decreased in AngII-infused MyD88-/- mice compared to C57BL/6 controls. Flow cytometric analyses revealed that AngII-induced aortic accumulation of CD11b+Ly6Chi inflammatory monocytes was significantly dampened in MyD88-/- mice. AngII increased mRNA expression of cd62L, cd68 and ccl2, which was blocked in MyD88-/- mice, indicating a role of MyD88 for myeloid cell activation and differentiation. Additionally, aortic interferon-g+ NK cells and mRNA levels of interleukin-12 and interleukin-1b were reduced in AngII-treated MyD88-/- mice. Bone marrow transfer experiments demonstrated a vasoprotective effect of MyD88 deficiency in bone marrow-derived cells, and selective expression of MyD88 in LysMCre/wtMyD88LSL/LSL mice restored AngII-induced pathology, revealing that myelomonocytic cells drives vascular dysfunction in a MyD88-dependent manner. Furthermore, data from 1,274 individuals in the GHS indicated, that MyD88 mRNA expression was associated with all cause mortality and incident chronic heart failure after a median follow-up of 11.6 years. Computational analysis of the human protein interaction network demonstrated that MyD88 is significantly connected to proteins encoded by genetic loci associated with blood pressure traits in multiple GWAS. Conclusion We provide evidence, that MyD88 expressed by myelomonocytic cells promotes AngII-induced vascular dysfunction and arterial hypertension. MyD88 might be used as an inflammatory diagnostic marker for the risk of death in hypertensive patients.

Unveiling Crucial Proteins in Zone 2 of the Human Protein Interaction Network: Implications for Cancer Research and Therapeutic Targeting

Background: Protein-protein interaction networks are of great importance for understanding the complexity of biological processes and diseases, including cancer. However, detecting key proteins that regulate and mediate biological processes within these networks can be challenging due to their enormous size and complexity. Identification of these important proteins is of great importance for identifying therapeutic targets and elucidating molecular mechanisms in various types of cancer. Aims: The aim of this study was to investigate the importance of zone 2, the central zone of the protein interaction network in humans, and to identify potential therapeutic targets for common proteins expressed in different types of cancer. Method: We analyze zone 2 of the human protein interaction network and identify proteins within the network that function as key regulators or mediators of biological processes. We use multiple biological databases to collect data on protein-protein interactions and cancer types. We analyze the data to obtain promising proteins for different types of cancer. Results: Our analysis identified several important proteins in several types of cancer. These proteins play important roles in regulating biological processes within the network. Conclusion: Our approach effectively identifies essential proteins within protein-protein interaction networks across multiple cancer types. Furthermore, the identification of key proteins in zone 2 of the human protein interaction network will provide new insights into the molecular mechanisms underlying various cancer types and pave the way for new therapeutic targets.

Decoding the host-pathogen interspecies molecular crosstalk during oral candidiasis in humans: an in silico analysis.

Introduction: The objective of this study is to investigate the interaction between Candida albicans and human proteins during oral candidiasis, with the aim of identifying pathways through which the pathogen subverts host cells. Methods: A comprehensive list of interactions between human proteins and C. albicans was obtained from the Human Protein Interaction Database using specific screening criteria. Then, the genes that exhibit differential expression during oral candidiasis in C. albicans were mapped with the list of human-Candida interactions to identify the corresponding host proteins. The identified host proteins were further compared with proteins specific to the tongue, resulting in a final list of 99 host proteins implicated in oral candidiasis. The interactions between host proteins and C. albicans proteins were analyzed using the STRING database, enabling the construction of protein-protein interaction networks. Similarly, the gene regulatory network of Candida proteins was reconstructed using data from the PathoYeastract and STRING databases. Core module proteins within the targeted host protein-protein interaction network were identified using ModuLand, a Cytoscape plugin. The expression levels of the core module proteins under diseased conditions were assessed using data from the GSE169278 dataset. To gain insights into the functional characteristics of both host and pathogen proteins, ontology analysis was conducted using Enrichr and YeastEnrichr, respectively. Result: The analysis revealed that three Candida proteins, HHT21, CYP5, and KAR2, interact with three core host proteins, namely, ING4 (in the DNMT1 module), SGTA, and TOR1A. These interactions potentially impair the immediate immune response of the host against the pathogen. Additionally, differential expression analysis of fungal proteins and their transcription factors in Candida-infected oral cell lines indicated that Rob1p, Tye7p, and Ume6p could be considered candidate transcription factors involved in instigating the pathogenesis of oral candidiasis during host infection. Conclusion: Our study provides a molecular map of the host-pathogen interaction during oral candidiasis, along with potential targets for designing regimens to overcome oral candidiasis, particularly in immunocompromised individuals.

Proteomic Analyses Reveal Functional Pathways and Potential Targets in Pediatric Hydrocephalus.

Hydrocephalus is a common pediatric disorder of cerebral spinal fluid physiology resulting in abnormal expansion of the cerebral ventricles. However, the underlying molecular mechanisms remain unknown. We performed proteomic analyses of cerebrospinal fluid (CSF) from 7 congenital hydrocephalus and 5 arachnoid cyst patients who underwent surgical treatment. Differentially expressed proteins (DEPs) were identified by label-free Mass Spectrometry followed by differential expression analysis. The GO and GSEA enrichment analysis was performed to explore the cancer hallmark pathways and immune-related pathways affected by DEPs. Then, network analysis was applied to reveal the location of DEPs in the human protein-protein interactions (PPIs) network. Potential drugs for hydrocephalus were identified based on drug-target interaction. We identified 148 up-regulated proteins and 82 down-regulated proteins, which are potential biomarkers for clinical diagnosis of hydrocephalus and arachnoid cyst. Functional enrichment analysis revealed that the DEPs were significantly enriched in the cancer hallmark pathways and immunerelated pathways. In addition, network analysis uncovered that DEPs were more likely to be located in the central regions of the human PPIs network, suggesting DEPs may be proteins that play important roles in human PPIs. Finally, we calculated the overlap of drug targets and the DEPs based on drugtarget interaction to identify the potential therapeutic drugs of hydrocephalus. The comprehensive proteomic analyses provided valuable resources for investigating the molecular pathways in hydrocephalus, and uncovered potential biomarkers for clinical diagnosis and therapy.

DapBCH: a disease association prediction model Based on Cross-species and Heterogeneous graph embedding.

The study of comorbidity can provide new insights into the pathogenesis of the disease and has important economic significance in the clinical evaluation of treatment difficulty, medical expenses, length of stay, and prognosis of the disease. In this paper, we propose a disease association prediction model DapBCH, which constructs a cross-species biological network and applies heterogeneous graph embedding to predict disease association. First, we combine the human disease-gene network, mouse gene-phenotype network, human-mouse homologous gene network, and human protein-protein interaction network to reconstruct a heterogeneous biological network. Second, we apply heterogeneous graph embedding based on meta-path aggregation to generate the feature vector of disease nodes. Finally, we employ link prediction to obtain the similarity of disease pairs. The experimental results indicate that our model is highly competitive in predicting the disease association and is promising for finding potential disease associations.

Total network controllability analysis discovers explainable drugs for Covid-19 treatment

BackgroundThe active pursuit of network medicine for drug repurposing, particularly for combating Covid-19, has stimulated interest in the concept of structural controllability in cellular networks. We sought to extend this theory, focusing on the defense rather than control of the cell against viral infections. Accordingly, we extended structural controllability to total structural controllability and introduced the concept of control hubs. Perturbing any control hub may render the cell uncontrollable by exogenous stimuli like viral infections, so control hubs are ideal drug targets.ResultsWe developed an efficient algorithm to identify all control hubs, applying it to a largest homogeneous network of human protein interactions, including interactions between human and SARS-CoV-2 proteins. Our method recognized 65 druggable control hubs with enriched antiviral functions. Utilizing these hubs, we categorized potential drugs into four groups: antiviral and anti-inflammatory agents, drugs acting on the central nervous system, dietary supplements, and compounds enhancing immunity. An exemplification of our approach’s effectiveness, Fostamatinib, a drug initially developed for chronic immune thrombocytopenia, is now in clinical trials for treating Covid-19. Preclinical trial data demonstrated that Fostamatinib could reduce mortality rates, ICU stay length, and disease severity in Covid-19 patients.ConclusionsOur findings confirm the efficacy of our novel strategy that leverages control hubs as drug targets. This approach provides insights into the molecular mechanisms of potential therapeutics for Covid-19, making it a valuable tool for interpretable drug discovery. Our new approach is general and applicable to repurposing drugs for other diseases.

Maximum likelihood estimation for reversible mechanistic network models.

Mechanistic network models specify the mechanisms by which networks grow and change, allowing researchers to investigate complex systems using both simulation and analytical techniques. Unfortunately, it is difficult to write likelihoods for instances of graphs generated with mechanistic models, and thus it is near impossible to estimate the parameters using maximum likelihood estimation. In this paper, we propose treating the node sequence in a growing network model as an additional parameter, or as a missing random variable, and maximizing over the resulting likelihood. We develop this framework in the context of a simple mechanistic network model, used to study gene duplication and divergence, and test a variety of algorithms for maximizing the likelihood in simulated graphs. We also run the best-performing algorithm on one human protein-protein interaction network and four nonhuman protein-protein interaction networks. Although we focus on a specific mechanistic network model, the proposed framework is more generally applicable to reversible models.

Molecular complex detection in protein interaction networks through reinforcement learning

BackgroundProteins often assemble into higher-order complexes to perform their biological functions. Such protein–protein interactions (PPI) are often experimentally measured for pairs of proteins and summarized in a weighted PPI network, to which community detection algorithms can be applied to define the various higher-order protein complexes. Current methods include unsupervised and supervised approaches, often assuming that protein complexes manifest only as dense subgraphs. Utilizing supervised approaches, the focus is not on how to find them in a network, but only on learning which subgraphs correspond to complexes, currently solved using heuristics. However, learning to walk trajectories on a network to identify protein complexes leads naturally to a reinforcement learning (RL) approach, a strategy not extensively explored for community detection. Here, we develop and evaluate a reinforcement learning pipeline for community detection on weighted protein–protein interaction networks to detect new protein complexes. The algorithm is trained to calculate the value of different subgraphs encountered while walking on the network to reconstruct known complexes. A distributed prediction algorithm then scales the RL pipeline to search for novel protein complexes on large PPI networks.ResultsThe reinforcement learning pipeline is applied to a human PPI network consisting of 8k proteins and 60k PPI, which results in 1,157 protein complexes. The method demonstrated competitive accuracy with improved speed compared to previous algorithms. We highlight protein complexes such as C4orf19, C18orf21, and KIAA1522 which are currently minimally characterized. Additionally, the results suggest TMC04 be a putative additional subunit of the KICSTOR complex and confirm the involvement of C15orf41 in a higher-order complex with HIRA, CDAN1, ASF1A, and by 3D structural modeling.ConclusionsReinforcement learning offers several distinct advantages for community detection, including scalability and knowledge of the walk trajectories defining those communities. Applied to currently available human protein interaction networks, this method had comparable accuracy with other algorithms and notable savings in computational time, and in turn, led to clear predictions of protein function and interactions for several uncharacterized human proteins.

Immunosensitivity mediated by downregulated AKT1-SKP2 induces anti-PD-1-associated thyroid immune injury.

Immune checkpoint inhibitors evoke the immune system, which may cause immune-related adverse effects. The predictors and mechanisms of anti-PD-1-associated thyroid immune injury remain unclear. A retrospective analysis is conducted on 518 patients treated with anti PD-1/PD-L1. Firstly, the differences between anti PD-1 and anti PD-L1 are compared on the risk of thyroid immune injury. Then, the predictors of the risk and thyroid function for anti PD-1 related thyroid immune injury are analyzed. Furthermore, the in vitro mechanism of normal thyroid cells (NTHY) is studied. First, the effect of anti PD-1 on the cell viability and immune sensitivity of thyroid cells is observed. Cell viability includes cell proliferation, apoptosis, cell cycle, T4 secretion, while immune sensitivity includes molecular expression and CD8+T cell aggregation and killing towards NTHY. Then the differentially expressed proteins (DEPs) are screened by protein mass spectrometry. Enrichment of KEGG pathway and annotation of GO function on DEPs are conducted. Human protein-protein interactions are obtained from STRING database. The network is constructed and analyzed using Cytoscape software. In vitro, key proteins and their pathways are validated through overexpression plasmids or inhibitors. The recovery experiment and the immuno-coprecipitation experiment are designed to support the results. In vivo, the key proteins are detected in the thyroid tissue of mice fed with anti PD-1, as well as in the thyroid tissue of patients with Hashimoto's thyroiditis. Thyroid irAE is associated with female, IgG, FT4, TPOAb, TGAb, TSHI, TFQI, and TSH. Peripheral lymphocytes are associated with thyroid function. In vitro, the NIVO group shows prologed G1 phase, decreased FT4, downregulated PD-L1, upregulated IFN-γ, and more CD8+T cell infiltration and cytotoxicity. AKT1-SKP2 is chosen as the key protein. AKT1 overexpression and SKP2 inhibitor replies to NIVO and AKT1 overexpression, respectively. Immunoprecipitation shows SKP2 and PD-L1 interaction. Female, impaired thyroid hormone sensitivity and IgG4 contribute to the risk of thyroid irAE, while peripheral blood lymphocyte characteristics affect thyroid function. Anti-PD-1 induces thyroid irAE by downregulating AKT1-SKP2 to enhance thyroid immunosensitivity.