Driven Prediction of Protein Interaction Outcomes
Cut Experimental Cost and Time Before Entering the Lab
Reduce experimental iterations by predicting binding affinity, stability, and interaction order behavior before wet-lab validation.
Reduce the number of experiments conducted from 10 to 4
Say goodbye to intermediate and verification experiments
in the laboratory!
in the laboratory!
Predict experimental outcomes for next research fields:
Experimental results:
Looking for antiviral therapy?
Then click here VIRUSES
- Binding/kinetics: SPR/BLI curves, KD/kon/koff, competition assays
- potency (IC50/EC50),
- pathway readouts,
- cell-based efficacy
- aggregation,
- thermal stability (DSF/DSC),
- solubility,
- Negative results, false/positive results,
- affinity optimization,
- specificity improvement,
- potency,
- stability,
- half-life,
- solubility,
- reduced aggregation,
Looking for antiviral therapy?
Then click here VIRUSES
-
Affinity optimization workflows
Machine learning predicts binding affinities (Kd, ΔG), residue-wise contributions, and recommends optimal mutations -
Antibody–antigen interactions
Stepwise affinity testing of each antibody–antigen pair, modification of antibody flexible chains guided by AI.
Engineering departments receive preliminary information on the composition and structure of high-affinity antibodies to the antigen. -
Protein–protein binding
Difficult protein expression/purification may introduce errors and produce false positive results.
In our laboratory, research into the prediction of the physical properties of proteins is carried out under ideal conditions. -
Commercial drug discovery R&D
with strong molecular and biochemical research. -
Biopharma R&D and development
with heavy protein/biochemical assay pipelines -
Biomanufacturing/process R&D
downstream purification for cost-effective, scalable, and safe production -
High-containment virology;
molecular/biochemical mechanisms of viral infection. -
Viral pathogenesis and viral immunology
fundamental cell/molecular virology. -
Major hub for virology
viral pathogenesis, vaccine/antibody research, and immune–virus biology -
Molecular mechanisms of viral infection
virus–host interactions; tools for discovery/diagnostics/therapeutics/vaccines -
Biochemistry, structural biology, molecular mechanism
strong protein/complex work and biophysical methods.proteins, signaling, enzymology, molecular mechanisms -
Hard link between static structures and function
-
Very large ligand/compound design space
-
Molecular medicine
with strong mechanistic biochemistry, signaling, systems biology -
Protein/biomolecular research and cellular biochemistry
-
Tumor & Microenvironment
-
Tumor immunology and therapy
-
Disease-driven molecular and biochemical/targets,
mechanisms, translational assays
How the Pilot Works?
Simple workflow and use of available laboratory data to start a pilot project
- Academic research laboratories
- Biotech R&D Teams
- Pharma Discovery Groups
- Core facilities
- Translational research groups
- Early-stage biotech spin-outs
Who This Is For:
What the Pilot Evaluates:
Antibody–antigen interactions:
Experimental condition ranking:
- optimization of antibodies to antigen
- Alanine scanning
- Mutagenesis
- Assembly order of protein complexes
- Key amino acid residues determination
- Stability of biochemical formation determination
- Kd, entropy change, dissociation energy determination
Experimental condition ranking:
- temperature, NaCl concentration
- Define a specific experimental question
- Upload sequences and optional structures
- Receive predicted interaction metrics
- Validate predictions experimentally
- Review correlation and performance report
How the Pilot Works:
- Academic research laboratories
- Biotech R&D Teams
- Pharma Discovery Groups
- Core facilities
- Translational research groups
- Early-stage biotech spin-outs
Who This Is For:
What the Pilot Evaluates:
Antibody–antigen interactions:
Experimental condition ranking:
- optimization of antibodies to antigen
- Alanine scanning
- Mutagenesis
- Assembly order of protein complexes
- Key amino acid residues determination
- Stability of biochemical formation determination
- Kd, entropy change, dissociation energy determination
Experimental condition ranking:
- temperature, NaCl concentration
- Define a specific experimental question
- Upload sequences and optional structures
- Receive predicted interaction metrics
- Validate predictions experimentally
- Review correlation and performance report
How the Pilot Works:
Detailed description with diagrams and examples. Examples of advanced features.
- modification of antibody flexible chains guided by AI,
- stepwise affinity testing of each antibody–antigen pair,
- identification of key binding residues for epitope mapping
- prediction of the affinity range (Kd, stability) based on structural input,
Affinity Prediction and Binding Analysis: Kd, ΔG, and Epitope Mapping
- Epitope Mapping: Alanine Scanning Mutagenesis,
- peptide-protein interactions
- identification of key binding residues
- prediction of the :
- affinity Kd, IC50% prognosis
- Stability parametr lg(cond(W))
- entropy change T(delta)S
- free energy change (delta)(delta) G
Protein–protein interaction : Kd, ΔG, and Epitope Mapping
- monomers [P1]
- dimers [P1-P1], [P1-P2]
- trimers [P1-P1-P1], [P1-P2-P3]
- protein+inhibitor [P1-Inh]
- dimer+inhibitors [P1-P2-Inh]
- tetramer+DNA
The correlation between calculated and
experimental parameters reaches 93%
experimental parameters reaches 93%
The prediction method is suitable for:
- Stability parametr lg(cond(W))
- entropy change T(delta)S
- free energy change (delta)(delta) G
Stability of protein structure under various experimental conditions (using various buffers)
Correlation between calculated lg(cond(W)) and
experimental data [urea]50% at the required increased denaturant concentration
experimental data [urea]50% at the required increased denaturant concentration
The maximum correlation dependence between the calculated and experimental values under conditions of p53 protein denaturation was found between lg(cond(W)) and the denaturant concentration [Urea]50% in the region of increased concentrations required for denaturation of 50% of the protein in solution. Starting with C1=2.8M and more.
The correlation dependence between the values reached 90%
The correlation dependence between the values reached 90%
List of p53 protein mutations, the denaturation of which requires an increased concentration of [Urea]50% denaturant
Protein stability prediction using deep learning
Analysis of the stability of protein structures upon the addition of denaturants.
An example of using the calculation of a monomeric protein using the example of P53 mutations. Calculations made it possible to separate neutral mutations from oncological mutations of the P53 protein.
Calculation results for negative and Positive set of P53 muttaions
Separation of oncogenic mutations from neutral mutations.
Systematized experimental pain points
Apply for the Pilot Project:
A Prognosis Approach to Antibody Escape and Virus–Host Binding Changes
A Prognosis Approach to Antibody Escape and Virus–Host Binding Changes
High-variability (fast adaptation) viruses:
HIV-1, influenza A, HCV, SARS-CoV-2
HIV-1, influenza A, HCV, SARS-CoV-2
Viruses where drug resistance is a defining feature:
HIV-1: drug resistance is a major public health issue.
Influenza A: resistance can arise through well-known mutations (e.g., H275Y is associated with strong reduction in susceptibility to oseltamivir).
SARS-CoV-2: resistance mutations can emerge during antiviral treatment, especially monitored with deep sequencing.
HSV / HCMV: resistance can occur (clinically important in immunocompromised patients), though at a different scale than HIV/influenza.
HIV-1: drug resistance is a major public health issue.
Influenza A: resistance can arise through well-known mutations (e.g., H275Y is associated with strong reduction in susceptibility to oseltamivir).
SARS-CoV-2: resistance mutations can emerge during antiviral treatment, especially monitored with deep sequencing.
HSV / HCMV: resistance can occur (clinically important in immunocompromised patients), though at a different scale than HIV/influenza.
- Variant-risk heatmap (antibody–virus) + host-binding shift map (virus–host)
- How viral mutations can also shift virus–host protein affinity (e.g., receptor / host-factor binding),
- to design a computational triage workflow that prioritizes which variants would be most informative to test—purely in theory.
- For each variant, estimate relative changes in binding (e.g., ΔKd or ΔΔG trend) and summarize which mutations drive the biggest changes.
- The assembly order and attachment of the virulent particle to the host RNA/DNA
What we do:
Determination Virus–host protein affinity shifts:
Determination Virus–host protein affinity shifts:
- Variant-risk heatmap (antibody–virus) + host-binding shift map (virus–host)
- How viral mutations can also shift virus–host protein affinity (e.g., receptor / host-factor binding),
- to design a computational triage workflow that prioritizes which variants would be most informative to test—purely in theory.
- For each variant, estimate relative changes in binding (e.g., ΔKd or ΔΔG trend) and summarize which mutations drive the biggest changes.
- The assembly order and attachment of the virulent particle to the host RNA/DNA
What we do:
Determination Virus–host protein affinity shifts:
Determination Virus–host protein affinity shifts:
| |
Start applying for the pilot project! Fill in the fields with the required information about your project. Check the box below the form if you require strict confidentiality. |
