Analysis of the Plasticity of Circulating Tumor Cells Reveals Differentially Regulated Kinases During the Suspension-to-Adherent Transition.

BACKGROUND: Research on circulating tumor cells (CTCs) offers the opportunity to better understand the initial steps of blood-borne metastasis as main cause of cancer-related deaths. Here, we have used the colon cancer CTC-MCC-41 and breast cancer CTC-ITB-01 lines, which were both established from human CTCs as permanent cell lines as models to further study CTC biology with special emphasis on anchorage-independent survival and growth. METHODS AND RESULTS: Both cell lines showed a marked intrinsic plasticity to switch between suspension and adherent in vitro growth, in 2D adherent culture conditions, and established an equilibrium of both growth patterns with predominant adherent cells in the CTC-MCC-41 line (77%) and suspension cells in the CTC-ITB-01 line (85%). Western blot analysis revealed a higher expression of pERK1/2 in CTC-ITB-01 adherent cells compared to the suspension counterpart that suggested the involvement of kinases in this process. Subsequent functional kinome profiling identified several serine/threonine as well as tyrosine kinases that were differentially regulated in adherent and suspension CTCs. In the adherent cells of the breast cancer line CTC-ITB-01 the activity of MSK1, Src family kinases and the PKG family was increased compared to the suspension counterpart. In adherent cells of the colorectal CTC-MCC-41 line, an increased activity of TYRO3 and JAK2 was detected, whereas p38 MAPK was strongly impaired in the suspension CTC-MCC-41 cells. Some of the regulated kinases, which include the Src family, TYRO3, MSK1, JAK2 and p38 MAPK, have been associated with crucial cellular processes including proliferation, migration and dormancy in the past. CONCLUSIONS: The investigated CTC lines exhibit a high plasticity, similar to the concept of 'adherent-to-suspension transition (AST)' that was recently suggested as a new hallmark of tumor biology by Huh et al. Moreover, we identified differentially regulated kinome profiles that may represent potential targets for future studies on therapeutic interventions.

CNN-BLSTM based deep learning framework for eukaryotic kinome classification: An explainability based approach.

Classification of protein families from their sequences is an enduring task in Proteomics and related studies. Numerous deep-learning models have been moulded to tackle this challenge, but due to the black-box character, they still fall short in reliability. Here, we present a novel explainability pipeline that explains the pivotal decisions of the deep learning model on the classification of the Eukaryotic kinome. Based on a comparative and experimental analysis of the most cutting-edge deep learning algorithms, the best deep learning model CNN-BLSTM was chosen to classify the eight eukaryotic kinase sequences to their corresponding families. As a substitution for the conventional class activation map-based interpretation of CNN-based models in the domain, we have cascaded the GRAD CAM and Integrated Gradient (IG) explainability modus operandi for improved and responsible results. To ensure the trustworthiness of the classifier, we have masked the kinase domain traces, identified from the explainability pipeline and observed a class-specific drop in F1-score from 0.96 to 0.76. In compliance with the Explainable AI paradigm, our results are promising and contribute to enhancing the trustworthiness of deep learning models for biological sequence-associated studies.

Impact of protein and small molecule interactions on kinase conformations.

Protein kinases act as central molecular switches in the control of cellular functions. Alterations in the regulation and function of protein kinases may provoke diseases including cancer. In this study we investigate the conformational states of such disease-associated kinases using the high sensitivity of the kinase conformation (KinCon) reporter system. We first track BRAF kinase activity conformational changes upon melanoma drug binding. Second, we also use the KinCon reporter technology to examine the impact of regulatory protein interactions on LKB1 kinase tumor suppressor functions. Third, we explore the conformational dynamics of RIP kinases in response to TNF pathway activation and small molecule interactions. Finally, we show that CDK4/6 interactions with regulatory proteins alter conformations which remain unaffected in the presence of clinically applied inhibitors. Apart from its predictive value, the KinCon technology helps to identify cellular factors that impact drug efficacies. The understanding of the structural dynamics of full-length protein kinases when interacting with small molecule inhibitors or regulatory proteins is crucial for designing more effective therapeutic strategies.

The Histogenetic Origin of Malignant Cells Predicts Their Susceptibility towards Synthetic Lethality Utilizing the TK.007 System.

BACKGROUND: Remarkable differences exist in the outcome of systemic cancer therapies. Lymphomas and leukemias generally respond well to systemic chemotherapies, while solid cancers often fail. We engineered different human cancer cells lines to uniformly express a modified herpes simplex virus thymidine kinase TK.007 as a suicide gene when ganciclovir (GCV) is applied, thus in theory achieving a similar response in all cell lines. METHODS: Fifteen different cell lines were engineered to express the TK.007 gene. XTT-cell proliferation assays were performed and the IC50-values were calculated. Functional kinome profiling, mRNA sequencing, and bottom-up proteomics analysis with Ingenuity pathway analysis were performed. RESULTS: GCV potency varied among cell lines, with lymphoma and leukemia cells showing higher susceptibility than solid cancer cells. Functional kinome profiling implies a contribution of the SRC family kinases and decreased overall kinase activity. mRNA sequencing highlighted alterations in the MAPK pathways and bottom-up proteomics showed differences in apoptotic and epithelial junction signaling proteins. CONCLUSIONS: The histogenetic origin of cells influenced the susceptibility of human malignant cells towards cytotoxic agents with leukemias and lymphomas being more sensitive than solid cancer cells.

Pneumococcal hydrogen peroxide regulates host cell kinase activity.

Introduction: Protein kinases are indispensable reversible molecular switches that adapt and control protein functions during cellular processes requiring rapid responses to internal and external events. Bacterial infections can affect kinase-mediated phosphorylation events, with consequences for both innate and adaptive immunity, through regulation of antigen presentation, pathogen recognition, cell invasiveness and phagocytosis. Streptococcus pneumoniae (Spn), a human respiratory tract pathogen and a major cause of community-acquired pneumoniae, affects phosphorylation-based signalling of several kinases, but the pneumococcal mediator(s) involved in this process remain elusive. In this study, we investigated the influence of pneumococcal H2O2 on the protein kinase activity of the human lung epithelial H441 cell line, a generally accepted model of alveolar epithelial cells. Methods: We performed kinome analysis using PamGene microarray chips and protein analysis in Western blotting in H441 lung cells infected with Spn wild type (SpnWT) or with SpnΔlctOΔspxB -a deletion mutant strongly attenuated in H2O2 production- to assess the impact of pneumococcal hydrogen peroxide (H2O2) on global protein kinase activity profiles. Results: Our kinome analysis provides direct evidence that kinase activity profiles in infected H441 cells significantly vary according to the levels of pneumococcal H2O2. A large number of kinases in H441 cells infected with SpnWT are significantly downregulated, whereas this no longer occurs in cells infected with the mutant SpnΔlctOΔspxB strain, which lacks H2O2. In particular, we describe for the first time H2O2-mediated downregulation of Protein kinase B (Akt1) and activation of lymphocyte-specific tyrosine protein kinase (Lck) via H2O2-mediated phosphorylation.

Translational Relevance of Secondary Intracellular Signaling Cascades Following Traumatic Spinal Cord Injury.

Traumatic spinal cord injury (SCI) is a life-threatening and life-altering condition that results in debilitating sensorimotor and autonomic impairments. Despite significant advances in the clinical management of traumatic SCI, many patients continue to suffer due to a lack of effective therapies. The initial mechanical injury to the spinal cord results in a series of secondary molecular processes and intracellular signaling cascades in immune, vascular, glial, and neuronal cell populations, which further damage the injured spinal cord. These intracellular cascades present promising translationally relevant targets for therapeutic intervention due to their high ubiquity and conservation across eukaryotic evolution. To date, many therapeutics have shown either direct or indirect involvement of these pathways in improving recovery after SCI. However, the complex, multifaceted, and heterogeneous nature of traumatic SCI requires better elucidation of the underlying secondary intracellular signaling cascades to minimize off-target effects and maximize effectiveness. Recent advances in transcriptional and molecular neuroscience provide a closer characterization of these pathways in the injured spinal cord. This narrative review article aims to survey the MAPK, PI3K-AKT-mTOR, Rho-ROCK, NF-κB, and JAK-STAT signaling cascades, in addition to providing a comprehensive overview of the involvement and therapeutic potential of these secondary intracellular pathways following traumatic SCI.

Sex-specific sequels of early life stress on serine/threonine kinase activity in visceral adipose tissue from obese mice.

Adverse childhood experiences (ACEs) are an established independent risk factor for chronic disease including obesity and hypertension; however, only women exposed to multiple ACEs show a positive relationship with BMI. Our lab has reported that maternal separation and early weaning (MSEW), a mouse model of early life stress, induces sex-specific mechanisms underlying greater blood pressure response to a chronic high fat diet (HF). Specifically, female MSEW mice fed a HF display exacerbated perigonadal white adipose tissue (pgWAT) expansion and a metabolic syndrome-like phenotype compared to control counterparts, whereas hypertension is caused by sympathoactivation in male MSEW mice. Thus, this study aimed to determine whether there is a sex-specific serine/threonine kinase (STKA) activity in pgWAT adipose tissue associated with early life stress. Frozen pgWAT was collected from MSEW and control, male and female mice fed a HF to assess STKA activity using the Pamstation12 instrument. Overall, MSEW induces significant reduction of 7 phosphokinases (|Z| >=1.5) in females (QIK, MLK, PKCH, MST, STE7, PEK, FRAY) and 5 in males (AKT, SGK, P38, MARK, CDK), while 15 were downregulated in both sexes (DMPK, PKA, PKG, RSK, PLK, DYRK, NMO, CAMK1, JNK, PAKA, RAD53, ERK, PAKB, PKD, PIM, AMPK). This data provides new insights into the sex-specific dysregulation of the molecular network controlling cellular phosphorylation signals in visceral adipose tissue and identifies possible target phosphokinases implicated in adipocyte hypertrophy as a result of exposure to early life stress. Identifying functional metabolic signatures is critical to elucidate the underlying molecular mechanisms behind the sex-specific obesity risk associated with early life stress.

Integrative Kinase Activity Profiling and Phosphoproteomics of rd10 Mouse Retina during cGMP-Dependent Retinal Degeneration.

Inherited retinal degenerative diseases (IRDs) are a group of rare diseases that lead to a progressive loss of photoreceptor cells and, ultimately, blindness. The overactivation of cGMP-dependent protein kinase G (PKG), one of the key effectors of cGMP-signaling, was previously found to be involved in photoreceptor cell death and was studied in murine IRD models to elucidate the pathophysiology of retinal degeneration. However, PKG is a serine/threonine kinase (STK) with several hundred potential phosphorylation targets and, so far, little is known about the specificity of the target interaction and downstream effects of PKG activation. Here, we carried out both the kinome activity and phosphoproteomic profiling of organotypic retinal explant cultures derived from the rd10 mouse model for IRD. After treating the explants with the PKG inhibitor CN03, an overall decrease in peptide phosphorylation was observed, with the most significant decrease occurring in seven peptides, including those from the known PKG substrate cyclic-AMP-response-element-binding CREB, but also Ca2+/calmodulin-dependent kinase (CaMK) peptides and TOP2A. The phosphoproteomic data, in turn, revealed proteins with decreased phosphorylation, as well as proteins with increased phosphorylation. The integration of both datasets identified common biological networks altered by PKG inhibition, which included kinases predominantly from the so-called AGC and CaMK families of kinases (e.g., PKG1, PKG2, PKA, CaMKs, RSKs, and AKTs). A pathway analysis confirmed the role of CREB, Calmodulin, mitogen-activated protein kinase (MAPK) and CREB modulation. Among the peptides and pathways that showed reduced phosphorylation activity, the substrates CREB, CaMK2, and CaMK4 were validated for their retinal localization and activity, using immunostaining and immunoblotting in the rd10 retina. In summary, the integrative analysis of the kinome activity and phosphoproteomic data revealed both known and novel PKG substrates in a murine IRD model. This data establishes a basis for an improved understanding of the biological pathways involved in cGMP-mediated photoreceptor degeneration. Moreover, validated PKG targets like CREB and CaMKs merit exploration as novel (surrogate) biomarkers to determine the effects of a clinical PKG-targeted treatment for IRDs.

Assessment of Kinome-Wide Activity Remodeling upon Picornavirus Infection.

Picornaviridae represent a large family of single-stranded positive RNA viruses of which different members can infect both humans and animals. These include the enteroviruses (e.g., poliovirus, coxsackievirus, and rhinoviruses) as well as the cardioviruses (e.g., encephalomyocarditis virus). Picornaviruses have evolved to interact with, use, and/or evade cellular host systems to create the optimal environment for replication and spreading. It is known that viruses modify kinase activity during infection, but a proteome-wide overview of the (de)regulation of cellular kinases during picornavirus infection is lacking. To study the kinase activity landscape during picornavirus infection, we here applied dedicated targeted mass spectrometry-based assays covering ∼40% of the human kinome. Our data show that upon infection, kinases of the MAPK pathways become activated (e.g., ERK1/2, RSK1/2, JNK1/2/3, and p38), while kinases involved in regulating the cell cycle (e.g., CDK1/2, GWL, and DYRK3) become inactivated. Additionally, we observed the activation of CHK2, an important kinase involved in the DNA damage response. Using pharmacological kinase inhibitors, we demonstrate that several of these activated kinases are essential for the replication of encephalomyocarditis virus. Altogether, the data provide a quantitative understanding of the regulation of kinome activity induced by picornavirus infection, providing a resource important for developing novel antiviral therapeutic interventions.

Comprehensive Data-Driven Assessment of Non-Kinase Targets of Inhibitors of the Human Kinome.

Protein kinases (PKs) are involved in many intracellular signal transduction pathways through phosphorylation cascades and have become intensely investigated pharmaceutical targets over the past two decades. Inhibition of PKs using small-molecular inhibitors is a premier strategy for the treatment of diseases in different therapeutic areas that are caused by uncontrolled PK-mediated phosphorylation and aberrant signaling. Most PK inhibitors (PKIs) are directed against the ATP cofactor binding site that is largely conserved across the human kinome comprising 518 wild-type PKs (and many mutant forms). Hence, these PKIs often have varying degrees of multi-PK activity (promiscuity) that is also influenced by factors such as single-site mutations in the cofactor binding region, compound binding kinetics, and residence times. The promiscuity of PKIs is often-but not always-critically important for therapeutic efficacy through polypharmacology. Various in vitro and in vivo studies have also indicated that PKIs have the potential of interacting with additional targets other than PKs, and different secondary cellular targets of individual PKIs have been identified on a case-by-case basis. Given the strong interest in PKs as drug targets, a wealth of PKIs from medicinal chemistry and their activity data from many assays and biological screens have become publicly available over the years. On the basis of these data, for the first time, we conducted a systematic search for non-PK targets of PKIs across the human kinome. Starting from a pool of more than 155,000 curated human PKIs, our large-scale analysis confirmed secondary targets from diverse protein classes for 447 PKIs on the basis of high-confidence activity data. These PKIs were active against 390 human PKs, covering all kinase groups of the kinome and 210 non-PK targets, which included other popular pharmaceutical targets as well as currently unclassified proteins. The target distribution and promiscuity of the 447 PKIs were determined, and different interaction profiles with PK and non-PK targets were identified. As a part of our study, the collection of PKIs with activity against non-PK targets and the associated information are made freely available.

Olanzapine's effects on hypothalamic transcriptomics and kinase activity.

Olanzapine is a second-generation antipsychotic that disrupts metabolism and is associated with an increased risk of type 2 diabetes. The hypothalamus is a key region in the control of whole-body metabolic homeostasis. The objective of the current study was to determine how acute peripheral olanzapine administration affects transcription and serine/threonine kinase activity in the hypothalamus. Hypothalamus samples from rats were collected following the pancreatic euglycemic clamp, thereby allowing us to study endpoints under steady state conditions for plasma glucose and insulin. Olanzapine stimulated pathways associated with inflammation, but diminished pathways associated with the capacity to combat endoplasmic reticulum stress and G protein-coupled receptor activity. These pathways represent potential targets to reduce the incidence of type 2 diabetes in patients taking antipsychotics.

AlignScape, displaying sequence similarity using self-organizing maps.

The current richness of sequence data needs efficient methodologies to display and analyze the complexity of the information in a compact and readable manner. Traditionally, phylogenetic trees and sequence similarity networks have been used to display and analyze sequences of protein families. These methods aim to shed light on key computational biology problems such as sequence classification and functional inference. Here, we present a new methodology, AlignScape, based on self-organizing maps. AlignScape is applied to three large families of proteins: the kinases and GPCRs from human, and bacterial T6SS proteins. AlignScape provides a map of the similarity landscape and a tree representation of multiple sequence alignments These representations are useful to display, cluster, and classify sequences as well as identify functional trends. The efficient GPU implementation of AlignScape allows the analysis of large MSAs in a few minutes. Furthermore, we show how the AlignScape analysis of proteins belonging to the T6SS complex can be used to predict coevolving partners.

Kinome and phosphoproteome reprogramming underlies the aberrant immune responses in critically ill COVID-19 patients.

SARS-CoV-2 infection triggers extensive host immune reactions, leading to severe diseases in certain individuals. However, the molecular basis underlying the excessive yet non-productive immune responses in severe COVID-19 remains incompletely understood. In this study, we conducted a comprehensive analysis of the peripheral blood mononuclear cell (PBMC) proteome and phosphoproteome in sepsis patients positive or negative for SARS-CoV-2 infection, as well as healthy subjects, using quantitative mass spectrometry. Our findings demonstrate dynamic changes in the COVID-19 PBMC proteome and phosphoproteome during disease progression, with distinctive protein or phosphoprotein signatures capable of distinguishing longitudinal disease states. Furthermore, SARS-CoV-2 infection induces a global reprogramming of the kinome and phosphoproteome, resulting in defective adaptive immune response mediated by the B and T lymphocytes, compromised innate immune responses involving the SIGLEC and SLAM family of immunoreceptors, and excessive cytokine-JAK-STAT signaling. In addition to uncovering host proteome and phosphoproteome aberrations caused by SARS-CoV-2, our work recapitulates several reported therapeutic targets for COVID-19 and identified numerous new candidates, including the kinases PKG1, CK2, ROCK1/2, GRK2, SYK, JAK2/3, TYK2, DNA-PK, PKCδ, and the cytokine IL-12.

Kinome-Wide Synthetic Lethal Screen Identifies PANK4 as a Modulator of Temozolomide Resistance in Glioblastoma.

Temozolomide (TMZ) represents the cornerstone of therapy for glioblastoma (GBM). However, acquisition of resistance limits its therapeutic potential. The human kinome is an undisputable source of druggable targets, still, current knowledge remains confined to a limited fraction of it, with a multitude of under-investigated proteins yet to be characterized. Here, following a kinome-wide RNAi screen, pantothenate kinase 4 (PANK4) isuncovered as a modulator of TMZ resistance in GBM. Validation of PANK4 across various TMZ-resistant GBM cell models, patient-derived GBM cell lines, tissue samples, as well as in vivo studies, corroborates the potential translational significance of these findings. Moreover, PANK4 expression is induced during TMZ treatment, and its expression is associated with a worse clinical outcome. Furthermore, a Tandem Mass Tag (TMT)-based quantitative proteomic approach, reveals that PANK4 abrogation leads to a significant downregulation of a host of proteins with central roles in cellular detoxification and cellular response to oxidative stress. More specifically, as cells undergo genotoxic stress during TMZ exposure, PANK4 depletion represents a crucial event that can lead to accumulation of intracellular reactive oxygen species (ROS) and subsequent cell death. Collectively, a previously unreported role for PANK4 in mediating therapeutic resistance to TMZ in GBM is unveiled.

FOXS1 is increased in liver fibrosis and regulates TGFß responsiveness and proliferation pathways in human hepatic stellate cells.

Liver fibrosis commences with liver injury stimulating transforming growth factor beta (TGFß) activation of hepatic stellate cells (HSCs), causing scarring and irreversible damage. TGFß induces expression of the transcription factor Forkhead box S1 (FOXS1) in hepatocytes and may have a role in the pathogenesis of hepatocellular carcinoma (HCC). To date, no studies have determined how it affects HSCs. We analyzed human livers with cirrhosis, HCC, and a murine fibrosis model and found that FOXS1 expression is significantly higher in fibrotic livers but not in HCC. Next, we treated human LX2 HSC cells with TGFß to activate fibrotic pathways, and FOXS1 mRNA was significantly increased. To study TGFß-FOXS1 signaling, we developed human LX2 FOXS1 CRISPR KO and scrambled control HSCs. To determine differentially expressed gene transcripts controlled by TGFß-FOXS1, we performed RNA-seq in the FOXS1 KO and control cells and over 400 gene responses were attenuated in the FOXS1 KO HSCs with TGFß-activation. To validate the RNA-seq findings, we used our state-of-the-art PamGene PamStation kinase activity technology that measures hundreds of signaling pathways nonselectively in real time. Using our RNA-seq data, kinase activity data, and descriptive measurements, we found that FOXS1 controls pathways mediating TGFß responsiveness, protein translation, and proliferation. Our study is the first to identify that FOXS1 may serve as a biomarker for liver fibrosis and HSC activation, which may help with early detection of hepatic fibrosis or treatment options for end-stage liver disease.

Development of new thieno[2,3-d]pyrimidines as dual EGFR and STAT3 inhibitors endowed with anticancer and pro-apoptotic activities.

In part due to the resilience of cellular feedback pathways that develop therapeutic resistance to targeting the EGFR alone, using EGFR inhibitors alone was demonstrated to be unsuccessful in clinical trials. The over-activation of the signal transducer/activator of transcription 3 (STAT3) during the administration of an EGFR inhibitor is expected to play a substantial part in the failure and resistance of EGFR inhibitor treatment. Therein, we proposed a hypothesis that induced STAT3-mediated resistance to EGFR inhibition therapy could be addressed by a dual inhibition of EGFR and STAT3 method. To this end, we tried to discover new thieno[2,3-d]pyrimidine derivatives "5a-o". Results from the screening on A549 and MCF7 cancer cell lines revealed that compounds 5j and 5k showed two-digit nanomolar with appropriate safety towards the WI-38 cell line. The best molecules, 5j and 5k, were subjected to γ-radiation, and their cytotoxic efficacy didn't change after irradiation, demonstrating that not having to use it avoided its side effects. Compounds 5j and 5k demonstrated the highest inhibition when their potency was tested as dual inhibitors on EGFR 67 and 41 nM, respectively, and STAT3 5.52 and 3.34 nM, respectively, proved with in silico molecular docking and dynamic simulation. In light of the results presented above, the capacity of both powerful compounds to alter the cell cycle and initiate the apoptotic process in breast cancer MCF7 cells was investigated. Caspase-8, Bcl-2, Bax and Caspase-9 apoptotic indicators were studied.

Illuminating function of the understudied druggable kinome.

The human kinome, with more than 500 proteins, is crucial for cell signaling and disease. Yet, about one-third of kinases lack in-depth study. The Data and Resource Generating Center for Understudied Kinases has developed multiple resources to address this challenge including creation of a heavy amino acid peptide library for parallel reaction monitoring and quantitation of protein kinase expression, use of understudied kinases tagged with a miniTurbo-biotin ligase to determine interaction networks by proximity-dependent protein biotinylation, NanoBRET probe development for screening chemical tool target specificity in live cells, characterization of small molecule chemical tools inhibiting understudied kinases, and computational tools for defining kinome architecture. These resources are available through the Dark Kinase Knowledgebase, supporting further research into these understudied protein kinases.

Variola Virus and Clade I Mpox Virus Differentially Modulate Cellular Responses Longitudinally in Monocytes During Infection.

Variola virus (VARV), the etiological agent of smallpox, had enormous impacts on global health prior to its eradication. In the absence of global vaccination programs, mpox virus (MPXV) has become a growing public health threat that includes endemic and nonendemic regions across the globe. While human mpox resembles smallpox in clinical presentation, there are considerable knowledge gaps regarding conserved molecular pathogenesis between these 2 orthopoxviruses. Thus, we sought to compare MPXV and VARV infections in human monocytes through kinome analysis. We performed a longitudinal analysis of host cellular responses to VARV infection in human monocytes as well as a comparative analysis to clade I MPXV-mediated responses. While both viruses elicited strong activation of cell responses early during infection as compared to later time points, several key differences in cell signaling events were identified and validated. These observations will help in the design and development of panorthopoxvirus therapeutics.

KinomeMETA: meta-learning enhanced kinome-wide polypharmacology profiling.

Kinase inhibitors are crucial in cancer treatment, but drug resistance and side effects hinder the development of effective drugs. To address these challenges, it is essential to analyze the polypharmacology of kinase inhibitor and identify compound with high selectivity profile. This study presents KinomeMETA, a framework for profiling the activity of small molecule kinase inhibitors across a panel of 661 kinases. By training a meta-learner based on a graph neural network and fine-tuning it to create kinase-specific learners, KinomeMETA outperforms benchmark multi-task models and other kinase profiling models. It provides higher accuracy for understudied kinases with limited known data and broader coverage of kinase types, including important mutant kinases. Case studies on the discovery of new scaffold inhibitors for membrane-associated tyrosine- and threonine-specific cdc2-inhibitory kinase and selective inhibitors for fibroblast growth factor receptors demonstrate the role of KinomeMETA in virtual screening and kinome-wide activity profiling. Overall, KinomeMETA has the potential to accelerate kinase drug discovery by more effectively exploring the kinase polypharmacology landscape.

KinScan: AI-based rapid profiling of activity across the kinome.

Kinases play a vital role in regulating essential cellular processes, including cell cycle progression, growth, apoptosis, and metabolism, by catalyzing the transfer of phosphate groups from adenosing triphosphate to substrates. Their dysregulation has been closely associated with numerous diseases, including cancer development, making them attractive targets for drug discovery. However, accurately predicting the binding affinity between chemical compounds and kinase targets remains challenging due to the highly conserved structural similarities across the kinome. To address this limitation, we present KinScan, a novel computational approach that leverages large-scale bioactivity data and integrates the Multi-Scale Context Aware Transformer framework to construct a virtual profiling model encompassing 391 protein kinases. The developed model demonstrates exceptional prediction capability, distinguishing between kinases by utilizing structurally aligned kinase binding site features derived from multiple sequence alignment for fast and accurate predictions. Through extensive validation and benchmarking, KinScan demonstrated its robust predictive power and generalizability for large-scale kinome-wide profiling and selectivity, uncovering associations with specific diseases and providing valuable insights into kinase activity profiles of compounds. Furthermore, we deployed a web platform for end-to-end profiling and selectivity analysis, accessible at https://kinscan.drugonix.com/softwares/kinscan.