EN

CN EN

AIDD/CADD

Home / CRO Services / Computational Solutions Integrating Dry and Wet Lab Approaches

Integrated Wet & Dry Lab Solutions

Leveraging the dual power of computation and experimentation, the Viva AIDD/CADD platform delivers integrated wet-and-dry lab solutions for diverse drug modalities. We provide an efficient optimization pathway at every stage of drug discovery by screening and refining molecules with multiple computational strategies. All results are validated through experimental methods, with feedback continuously used to optimize our AI models and strategies.

Computational Solutions for RNA-Targeting Small Molecules
  • Advanced RNA structure modeling and RNA-small molecule interaction simulation
  • Development of dedicated force fields for RNA and its non-canonical binding modes
  • Virtual screening of small molecules in a constrained chemical space (flatter and more polar RNA binding pockets)
Learn more
RNA-Targeting Small Molecule Discovery Workflow
  • System preparation

    Determination of simulation system atoms

    Force field parameter preparation for each component

    MD simulation configuration

  • Binding Pose Prediction

    Utilizing RNA molecular dynamics simulation to understand the mechanism of action (MOA)

    Establishing a method for predicting binding sites and poses

  • Binding affinity prediction

    Establish robust correlations between binding affinity and key metrics derived from MD trajectories.

  • Pilot MD simulation

    Understanding system dynamics

    Refining MD simulation conditions

  • Improved MD simulation

    Structure-based mechanism of action (MOA) study

    Continuous improvement of simulation conditions

  • Rational design

    Integrating known information into new compound design

    Using the established metric-affinity relationship to evaluate binding poses and affinity

  • System preparation

    Determination of simulation system atoms

    Force field parameter preparation for each component

    MD simulation configuration

  • Pilot MD simulation

    Understanding system dynamics

    Refining MD simulation conditions

  • Improved MD simulation

    Structure-based mechanism of action (MOA) study

    Continuous improvement of simulation conditions

  • Binding Pose Prediction

    Utilizing RNA molecular dynamics simulation to understand the mechanism of action (MOA)

    Establishing a method for predicting binding sites and poses

  • Binding affinity prediction

    Establish robust correlations between binding affinity and key metrics derived from MD trajectories.

  • Rational design

    Integrating known information into new compound design

    Using the established metric-affinity relationship to evaluate binding poses and affinity

Core Capabilities
  • Computational system characterization

    Comprehensive RNA-target structural analysis

    Identification of druggable binding sites

    Unveiling binding dynamics and thermodynamics

  • Quantitative affinity assessment and prediction

    High-precision binding affinity calculations

    Interaction analysis based on enhanced MD simulations

    RNA-target specificity assessment

  • Mechanism of action (MoA) understanding

    RNA conformational exploration

    Understanding of ligand recognition mechanisms

    Allosteric regulation analysis

  • Structure-based RNA-targeted drug design

    AI-driven ligand design

    AI-enhanced virtual screening

    AI/ML model-based developability prediction and optimization

Proprietary Advantages
  • Capturing system dynamics with enhanced MD sampling
    Overcoming the limitations of traditional methods that fail to capture the dynamic properties of RNA
  • No prior information required
    Breaking the limitations of traditional computational strategies that rely on structural information to define constraint parameters
  • Dedicated force fields
    Reducing errors caused by general-purpose force fields (e.g., inaccurate binding pose prediction)
  • Structure-based interaction analysis strategy
    Improving upon the poor correlation between general scoring systems and binding affinity
Integrated Wet-and-Dry Lab Smart Antibody Design Solutions
Providing a one-stop computational solution for "antibody discovery - affinity optimization - developability assessment"
  • End-to-end computation-driven
  • Protein-protein interaction (PPI)
  • Leading AI+physics models
  • Iterative optimization combined with experimental data
Antibodyde novoDesign
Antibody Humanization
Affinity Maturation
Developability Optimization
  • AI Model for Precise Structure Prediction
    Predicting antibody heavy and light chain structures with cutting-edge AI models
  • Dynamic Conformational Sampling
    Analyzing the flexible variable region of antibodies through enhanced molecular dynamics simulations and various sampling methods
  • Efficient Optimization of Antibody Developability
    Predicting and optimizing antibody developability
Case: de novo antibody design for novel targets
Performing MD simulations and analyzing the interaction networks of antibodies with known structures; the antibodies designed based on this approach retain key interactions.
  • Intelligent Humanization Design
    Utilizing deep learning and homology analysis to identify and optimize non-human antibody sequences
  • Precise Affinity Enhancement
    Identifying key mutation sites through FEP, QSAR, and AI-enhanced screening
  • Reduced Humanization Failure Risk
    Multi-objective sequence optimization by incorporating AI/ML prediction models for binding properties and developability
Case: computationally recommended humanization models based on murine antibodies

Through the antibody humanization solution on the VIVA Antibody Design Platform, we redesign murine antibodies, enabling them to maintain activity at both the molecular and cellular levels.



Four delivered humanized antibodies retained activity at the molecular level, with one of them maintaining activity at both the molecular and cellular levels.

Close to 100% success rate in humanization and affinity maturation.

Case: patent busting through computational antibody affinity maturation

Multi-target affinity fine-tuning for diverse requirements

Patent busting: maintain high affinity with CDR mutations

×
  • AI-assisted sequence optimization
  • Enhanced molecular dynamics simulation
  • Analysis of antibody surface charge and hydrophobicity
  • Immunogenicity and aggregation risk assessment
  • Comprehensive antibody developability assessment
  • Integrated optimization of computational protocols with experimental data
Case: antibody developability analysis based on sequence and structure
Case: antibody selectivity modulation
Case: antibody developability analysis based on sequence and structure

Assessing risks of antibody oxidation, deamidation, proteolysis, glycosylation, and isomerization

Customized improvement of physicochemical properties (e.g., viscosity, aggregation, solubility, and colloidal behavior)

Case: antibody selectivity modulation

Designing mutations to boost affinity for Antigen A and minimize the affinity gap with Antigen B in a bispecific antibody.

Peptide Computational Services with Experimental Validation
Ensuring the high efficiency and reliability of peptide optimization through a "computational design - experimental validation - optimization iteration" workflow.
  • AI-driven peptide design and optimization
  • Breaking the limitations of traditional druggable targets
  • Accelerating the discovery cycle
  • End-to-end one-stop solution
Learn more
Target to Hit
AI-driven DEL peptide screening

AI-assisted construction of focused peptides DEL libraries

DEL hit extraction and refinement combined with FEP calculation

Peptide virtual screening

Protein-peptide docking

Molecular dynamic simulation

Binding affinity calculation

Pharmacophore modeling

Similarity searching

2D/3D-QSAR modeling

de novo peptide design

Pocket-based de novo generation

Anchor-based cyclic peptide generation

Hit to Lead
Peptide rational design

Peptide structure prediction

Dynamic behavior of protein-peptide complexes analysis

Amino acids modification

Physics-based conformational sampling of unbound/bound peptide

Diverse cyclization strategy

Disulfide cyclization

Thioether bonds cyclization

Hydrocarbon stapling

Click chemistry cyclization

Binding affinity and stability evaluation

Polymer conjugation

AI-based peptide pharmacokinetics prediction
Permeability prediction

Caco2 regression model

PAMPA regression model

RRCK regression model

Stability prediction

t 1/2 in simulated gastric fluid classification model

t 1/2 in simulated intestinal fluid classification model

AI-driven Computational Solutions for PROTACs and Molecular Glues
  • AI-driven PROTAC/MG design

    AI-generated linkers

    AI-enhanced virtual screening: Based on Viva's proprietary linker library

  • Protein-Protein Interaction (PPI) analysis

    Leading proprietary PPI scoring system

    Optimization based on AI models and MD simulations

    Comprehensive judgment based on E2 and ubiquitin

  • Protein binding partners beyond E3 ligases

    Optimization for diverse binding partners

  • Enhancing strategy with Viva's exclusive PROTAC library

    Leverage Viva's experience in binder and linker synthesis for iterative design

Learn more

  • Identify target protein/E3 ligase pocket

    Screen, design and optimize the binder

    Design handle groups on the linker

  • Identify appropriate anchor points

    Explore linker length, flexibility and group type

    Improve the linker physiochemical property

  • Model ternary complex structure

    Evaluate PROTAC/MG ternary stability

    SAR analysis

  • Degradation activity

    Cell permeability

    PK/PD & ADMET properties

General Workflow for Virtual Linker Optimization
Case: Antibody Linker Site Selection
Determining linker structures with different lengths for Positions1 and Positions2 by analyzing complex conformations, dynamic behavior, and key metrics
Core Capabilities
  • Viva in-stock linker compound library
  • Ternary complex prediction
  • Protein-protein interaction(PPI) modeling
  • PROTAC linker screen & anchor point selection
  • Binder design
  • Linker de novo design
  • Druggability analysis
  • …….
Proprietary Advantages
  • Protein-protein interactions(PPI)

    Top-performing scoring system

    AI & MD optimization: post-docking refinement

    Enhanced PPI model: incorporates E2 and ubiquitin

  • For MG optimization beyond E3 ligase: targeting multiple binding partner
  • Expertise integration: leverage Viva's experience in binder & linker synthesis for iterative design
Case: AI-Driven discovery of E3 ligase binders
AIDD/CADD Platform
Online Message
After receiving your message, our professional team will contact you shortly.
  • Area of Interest

    CRO Drug R&D Services

    CDMO R&D and Production Services

    EFS Investment＆Incubation Business

    Media Cooperation

    Investor Relations

    Other

    How Did You Hear About Us?

    Search Engine (Google, Bing, Yahoo, etc)

    Conferences/Events

    Social Media

    LinkedIn

    Word of Mouth

    Other

×
Quick Link
Follow Us
Copyright Notice | Privacy Notice | Cookies Policy

Copyright © Viva Biotech All Rights Reserved. 沪ICP备19036061号