Inhibitors Virtual Screening And Designing For Three Important Target Proteins

Drug research and development is a complex and lengthy systematic project.Drug safety is of utmost importance during the development process,with its effectiveness and safety often dependent on selectivity for a specific target.Targets within the same family or type typically exhibit high homology,while their distribution in the human body may be tissue or cell specific.Hence,the selectivity of drug molecules for targets plays a crucial role in drug development.Researchers are diligently working towards developing drugs with enhanced selectivity to improve efficacy and minimize side effects.Recent research has highlighted the close relationship between most cancers and kinase targets,as well as ion channel targets.Furthermore,certain multifactorial cancers require simultaneous inhibition of multiple targets to achieve more effective therapeutic outcomes.As a result,this study explored kinase targets,ion channel targets,and dual-target selective inhibitors.Traditional drug design methods often encounter challenges like lengthy cycles,high costs,and significant risks.In recent years,the field has seen a shift towards leveraging the abundance of biological,chemical,and pharmaceutical data,coupled with advancements in computing power and algorithms,to focus on computer-aided drug research and development.With robust data processing,classification,and learning capabilities,computer-based approaches have shown promise in expediting drug optimization processes and enhancing overall research and development efficiency.This paper employs deep learning-based molecular generation and screening,receptor pocket-based molecular generation and screening methods,and machine learning-integrated screening methods to design and screen selective inhibitors for three important target proteins.The identified potential inhibitors are further validated through molecular dynamics simulations and binding free energy calculations.The main research contents were as follows:1 Long short-term memory Encoder-Decoder combined with fragment drug design method to discover RET selective inhibitorsThe receptor tyrosine kinase RET(Rearranged During Transfection)plays a crucial role in various cell signaling pathways and is essential for nervous system development.Abnormal activation of RET can lead to different types of cancers,such as thyroid cancer and non-small cell lung cancer.However,most RET inhibitors are multi-kinase inhibitors,posing a significant challenge in discovering inhibitors specifically targeting RET.In this study,a molecular generative model was developed by integrating the long short-term memory encoder-decoder with the fragment-based drug design(FBDD)method to create a molecular database for RET targets.Through molecular docking,molecular dynamics simulations,and binding free energy calculations,two potential selective inhibitors of RET were identified.These molecules were found to bind to the RET protein pocket in a specific conformation,demonstrating their specificity.This research offers a novel idea for selecting selective inhibitors for RET kinase targets.2 Design and screening bisamide scaffold inhibitors of TASK-1 potassium channelThe potassium channel known as TWIK-related Acid-sensitive K~+Channel 1(TASK-1)is extensively present in various tissues and plays a crucial role in neural activity and regulation of anesthesia.The dysregulated function of this channel has been linked to cancer,making it a significant target for drug research.Given the high similarity of 58.9%in human amino acid sequences between TASK-1 and TASK-3within the TASK family,achieving selectivity between these two channels poses a significant challenge.Previous studies on selective inhibitors of TASK-1 have shown that compounds with a bisamide scaffold exhibit some selectivity.In this study,we employed structure-based drug design(SBDD)and ligand-based drug design(LBDD)approaches to design TASK-1 channel selective inhibitors.By creating a molecular dataset centered around a bis-amide fragment as the core scaffold,utilizing molecular docking,3D-quantitative structure-activity relationship(3D-QSAR)analysis,molecular dynamics simulation,and incorporating binding free energy calculations,we successfully identified two potential selective inhibitor molecules for TASK-1 from the dataset.This research introduces a novel insight into developing and refining selective inhibitors targeting the TASK-1 ion channel.3 Combining machine learning with molecular dynamics simulations to discover dual-target inhibitors against BTK and JAK3The“one drug,one target”theory has successfully treated specific targets,but falls short when dealing with complex diseases like immune disorders and cancers.To effectively combat multifactorial diseases,it is essential to find drugs that can manipulate multiple targets simultaneously.By leveraging the synergistic effects of Bruton’s Tyrosine Kinase(BTK)and Janus kinase 3(JAK3),both critical targets in autoimmune diseases and hematologic malignancies,the simultaneous inhibition of BTK-JAK3 shows promise in treating B-cell lymphoma.In order to treat multifactorial diseases as well as off-target side effects,the discovery of inhibitors explicitly targeting the dual BTK-JAK3 target is crucial.This study utilized machine learning techniques,molecular docking,dynamics simulations,and binding free energy calculations to identify potential inhibitor molecules targeting BTK-JAK3.These molecules contain active fragments from BTK and JAK3,binding to both targets in a specific conformation,thus making them specific to the BTK-JAK3 dual target.The findings offer a screening process for developing selective inhibitors against dual targets and lay a theoretical foundation for future molecular optimization.