Machine Learning Modeling of Protein-intrinsic Features Predicts Tractability of Targeted Protein Degradation

Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development

Despite increasing interest in using Artificial Intelligence (AI) and Machine Learning (ML) models for drug development, effectively interpreting their predictions remains a challenge, which limits their impact on clinical decisions. We address this issue by providing a practical guide to SHapley Additive exPlanations (SHAP), a popular feature-based interpretability method, which can be seamlessly integrated into supervised ML models to gain a deeper understanding of their predictions, thereby enhancing their transparency and trustworthiness. This tutorial focuses on the application of SHAP analysis to standard ML black-box models for regression and classification problems. We provide an overview of various visualization plots and their interpretation, available software for implementing SHAP, and highlight best practices, as well as special considerations, when dealing with binary endpoints and time-series models. To enhance the reader's understanding for the method, we also apply it to inherently explainable regression models. Finally, we discuss the limitations and ongoing advancements aimed at tackling the current drawbacks of the method.

Mechanism of degrader-targeted protein ubiquitinability

Small-molecule degraders of disease-driving proteins offer a clinically proven modality with enhanced therapeutic efficacy and potential to tackle previously undrugged targets. Stable and long-lived degrader-mediated ternary complexes drive fast and profound target degradation; however, the mechanisms by which they affect target ubiquitination remain elusive. Here, we show cryo-EM structures of the VHL Cullin 2 RING E3 ligase with the degrader MZ1 directing target protein Brd4BD2 toward UBE2R1-ubiquitin, and Lys456 at optimal positioning for nucleophilic attack. In vitro ubiquitination and mass spectrometry illuminate a patch of favorably ubiquitinable lysines on one face of Brd4BD2, with cellular degradation and ubiquitinomics confirming the importance of Lys456 and nearby Lys368/Lys445, identifying the "ubiquitination zone." Our results demonstrate the proficiency of MZ1 in positioning the substrate for catalysis, the favorability of Brd4BD2 for ubiquitination by UBE2R1, and the flexibility of CRL2 for capturing suboptimal lysines. We propose a model for ubiquitinability of degrader-recruited targets, providing a mechanistic blueprint for further rational drug design.

Computational methods and key considerations for in silico design of proteolysis targeting chimera (PROTACs)

Proteolysis-targeting chimeras (PROTACs), as heterobifunctional molecules, have garnered significant attention for their ability to target previously undruggable proteins. Due to the challenges in obtaining crystal structures of PROTAC molecules in the ternary complex, a plethora of computational tools have been developed to aid in PROTAC design. These computational tools can be broadly classified into artificial intelligence (AI)-based or non-AI-based methods. This review aims to provide a comprehensive overview of the latest computational methods for the PROTAC design process, covering both AI and non-AI approaches, from protein selection to ternary complex modeling and prediction. Key considerations for in silico PROTAC design are discussed, along with additional considerations for deploying AI-based models. These considerations are intended to guide subsequent model development in the PROTAC design process. Finally, future directions and recommendations are provided.

A Kinetic Scout Approach Accelerates Targeted Protein Degrader Development

Bifunctional molecules such as targeted protein degraders induce proximity to promote gain-of-function pharmacology. These powerful approaches have gained broad traction across academia and the pharmaceutical industry, leading to an intensive focus on strategies that can accelerate their identification and optimization. We and others have previously used chemical proteomics to map degradable target space, and these datasets have been used to develop and train multiparameter models to extend degradability predictions across the proteome. In this study, we now turn our attention to develop generalizable chemistry strategies to accelerate the development of new bifunctional degraders. We implement lysine-targeted reversible-covalent chemistry to rationally tune the binding kinetics at the protein-of-interest across a set of 25 targets. We define an unbiased workflow consisting of global proteomics analysis, IP/MS of ternary complexes and the E-STUB assay, to mechanistically characterize the effects of ligand residence time on targeted protein degradation and formulate hypotheses about the rate-limiting step of degradation for each target. Our key finding is that target residence time is a major determinant of degrader activity, and this can be rapidly and rationally tuned through the synthesis of a minimal number of analogues to accelerate early degrader discovery and optimization efforts.

Lysineless HiBiT and NanoLuc Tagging Systems as Alternative Tools for Monitoring Targeted Protein Degradation

Target protein degradation (TPD) has emerged as a revolutionary approach in drug discovery, leveraging the cell's intrinsic machinery to selectively degrade disease-associated proteins. Nanoluciferase (nLuc) fusion proteins and the NanoBiT technology offer two robust and sensitive screening platforms to monitor the subtle changes in protein abundance induced by TPD molecules. Despite these advantages, concerns have arisen regarding potential degradation artifacts introduced by tagging systems due to the presence of lysine residues on them, prompting the development of alternative tools. In this study, we introduce HiBiT-RR and nLucK0, variants devoid of lysine residues, to mitigate such artifacts. Our findings demonstrate that HiBiT-RR maintains a similar sensitivity and binding affinity with the original HiBiT. Moreover, the comparison between nLucWT and nLucK0 constructs reveals variations in degradation patterns induced by certain TPD molecules, emphasizing the importance of choosing appropriate tagging systems to ensure the reliability of experimental outcomes in studying protein degradation processes.

P-NADs: PUX-based NAnobody degraders for ubiquitin-independent degradation of target proteins

Targeted protein degradation (TPD) allows cells to maintain a functional proteome and to rapidly adapt to changing conditions. Methods that repurpose TPD for the deactivation of specific proteins have demonstrated significant potential in therapeutic and research applications. Most of these methods are based on proteolysis targeting chimaeras (PROTACs) which link the protein target to an E3 ubiquitin ligase, resulting in the ubiquitin-based degradation of the target protein. In this study, we introduce a method for ubiquitin-independent TPD based on nanobody-conjugated plant ubiquitin regulatory X domain-containing (PUX) adaptor proteins. We show that the PUX-based NAnobody Degraders (P-NADs) can unfold a target protein through the Arabidopsis and human orthologues of the CDC48 unfoldase without the need for ubiquitination or initiating motifs. We demonstrate that P-NAD plasmids can be transfected into a human cell line, where the produced P-NADs use the endogenous CDC48 machinery for ubiquitin-independent TPD of a 143 kDa multidomain protein. Thus, P-NADs pave the road for ubiquitin-independent therapeutic TPD approaches. In addition, the modular P-NAD design combined with in vitro and cellular assays provide a versatile platform for elucidating functional aspects of CDC48-based TPD in plants and animals.

Targeting fusion oncoproteins in childhood cancers: challenges and future opportunities for developing therapeutics

Fusion oncoproteins are associated with childhood cancers and have proven challenging to target, aside from those that include kinases. As part of its efforts for targeting childhood cancers, the National Cancer Institute recently conducted a series on Novel Chemical Approaches for Targeting Fusion Oncoproteins. Key learnings on leading platforms and technologies that can be used to advance the development of molecular therapeutics that target fusion oncoproteins in childhood cancers are described. Recent breakthroughs in medicinal chemistry and chemical biology provide new ground and creative strategies to exploit for the development of targeted agents for improving outcomes against these recalcitrant cancers.

Artificial Intelligence in Point-of-Care Biosensing: Challenges and Opportunities

The integration of artificial intelligence (AI) into point-of-care (POC) biosensing has the potential to revolutionize diagnostic methodologies by offering rapid, accurate, and accessible health assessment directly at the patient level. This review paper explores the transformative impact of AI technologies on POC biosensing, emphasizing recent computational advancements, ongoing challenges, and future prospects in the field. We provide an overview of core biosensing technologies and their use at the POC, highlighting ongoing issues and challenges that may be solved with AI. We follow with an overview of AI methodologies that can be applied to biosensing, including machine learning algorithms, neural networks, and data processing frameworks that facilitate real-time analytical decision-making. We explore the applications of AI at each stage of the biosensor development process, highlighting the diverse opportunities beyond simple data analysis procedures. We include a thorough analysis of outstanding challenges in the field of AI-assisted biosensing, focusing on the technical and ethical challenges regarding the widespread adoption of these technologies, such as data security, algorithmic bias, and regulatory compliance. Through this review, we aim to emphasize the role of AI in advancing POC biosensing and inform researchers, clinicians, and policymakers about the potential of these technologies in reshaping global healthcare landscapes.

Lysineless HiBiT and NanoLuc Tagging Systems as Alternative Tools Monitoring Targeted Protein Degradation

Target protein degradation (TPD) has emerged as a revolutionary approach in drug discovery, leveraging the cell's intrinsic machinery to selectively degrade disease-associated proteins. Proteolysis-Targeting Chimeras (PROTACs) exemplify this strategy, exploiting heterobifunctional molecules to induce ubiquitination and subsequent degradation of target proteins. The clinical advancement of PROTACs underscores their potential in therapeutic intervention, with numerous projects progressing through clinical stages. However, monitoring subtle changes in protein abundance induced by TPD molecules demands highly sensitive assays. Nano-luciferase (nLuc) fusion proteins, or the NanoBiT technology derived from it, offer a robust screening platform due to their high sensitivity and stability. Despite these advantages, concerns have arisen regarding potential degradation artifacts introduced by tagging systems due to the presence of lysine residues on them, prompting the development of alternative tools. In this study, we introduce HiBiT-RR and nLuc K0 , variants devoid of lysine residues, to mitigate such artifacts. Our findings demonstrate that HiBiT-RR maintains similar sensitivity and binding affinity with the original HiBiT. Moreover, the comparison between nLuc WT and nLuc K0 constructs reveals variations in degradation patterns induced by certain PROTAC molecules, emphasizing the importance of choosing appropriate tagging systems to ensure the reliability of experimental outcomes in studying protein degradation processes.

Revolutionizing Drug Targeting Strategies: Integrating Artificial Intelligence and Structure-Based Methods in PROTAC Development

PROteolysis TArgeting Chimera (PROTAC) is an emerging technology in chemical biology and drug discovery. This technique facilitates the complete removal of the target proteins that are "undruggable" or challenging to target through chemical molecules via the Ubiquitin-Proteasome System (UPS). PROTACs have been widely explored and outperformed not only in cancer but also in other diseases. During the past few decades, several academic institutes and pharma companies have poured more efforts into PROTAC-related technologies, setting the stage for several major degrader trial readouts in clinical phases. Despite their promising results, the formation of robust ternary orientation, off-target activity, poor permeability, and binding affinity are some of the limitations that hinder their development. Recent advancements in computational technologies have facilitated progress in the development of PROTACs. Researchers have been able to utilize these technologies to explore a wider range of E3 ligases and optimize linkers, thereby gaining a better understanding of the effectiveness and safety of PROTACs in clinical settings. In this review, we briefly explore the computational strategies reported to date for the formation of PROTAC components and discuss the key challenges and opportunities for further research in this area.

Expanding PROTACtable genome universe of E3 ligases

Proteolysis-targeting chimera (PROTAC) and other targeted protein degradation (TPD) molecules that induce degradation by the ubiquitin-proteasome system (UPS) offer new opportunities to engage targets that remain challenging to be inhibited by conventional small molecules. One fundamental element in the degradation process is the E3 ligase. However, less than 2% amongst hundreds of E3 ligases in the human genome have been engaged in current studies in the TPD field, calling for the recruiting of additional ones to further enhance the therapeutic potential of TPD. To accelerate the development of PROTACs utilizing under-explored E3 ligases, we systematically characterize E3 ligases from seven different aspects, including chemical ligandability, expression patterns, protein-protein interactions (PPI), structure availability, functional essentiality, cellular location, and PPI interface by analyzing 30 large-scale data sets. Our analysis uncovers several E3 ligases as promising extant PROTACs. In total, combining confidence score, ligandability, expression pattern, and PPI, we identified 76 E3 ligases as PROTAC-interacting candidates. We develop a user-friendly and flexible web portal ( https://hanlaboratory.com/E3Atlas/ ) aimed at assisting researchers to rapidly identify E3 ligases with promising TPD activities against specifically desired targets, facilitating the development of these therapies in cancer and beyond.

Protein-protein interfaces in molecular glue-induced ternary complexes: classification, characterization, and prediction

Molecular glues are a class of small molecules that stabilize the interactions between proteins. Naturally occurring molecular glues are present in many areas of biology where they serve as central regulators of signaling pathways. Importantly, several clinical compounds act as molecular glue degraders that stabilize interactions between E3 ubiquitin ligases and target proteins, leading to their degradation. Molecular glues hold promise as a new generation of therapeutic agents, including those molecular glue degraders that can redirect the protein degradation machinery in a precise way. However, rational discovery of molecular glues is difficult in part due to the lack of understanding of the protein-protein interactions they stabilize. In this review, we summarize the structures of known molecular glue-induced ternary complexes and the interface properties. Detailed analysis shows different mechanisms of ternary structure formation. Additionally, we also review computational approaches for predicting protein-protein interfaces and highlight the promises and challenges. This information will ultimately help inform future approaches for rational molecular glue discovery.