Computational Models Accurately Predict Multi-Cell Biomarker Profiles in Inflammation and Cancer.

Area of Science:

• Computational biology
• Immunology
• Bioinformatics

Background:

• Inflammation and cancer involve complex cellular interactions influencing tissue microenvironments.
• Predicting collective cellular responses requires sophisticated modeling approaches.

Purpose of the Study:

• To develop and validate multi-cell computational models for predicting cellular biomarker profiles in inflamed or cancerous tissues.
• To assess the accuracy of these models in simulating chemokine and cytokine responses.
• To explore the potential of personalized computational models for therapeutic target identification.

Main Methods:

• Creation of individual cell computational models (myeloid, lymphoid, epithelial, cancer cells).
• Integration into multi-cell computational models to simulate tissue environments.
• Validation of model predictions against experimental data from cell cultures exposed to stimuli like LPS or Pam3CSK4.
• Incorporation of cell line-specific genomic data for personalized model simulations.

Main Results:

• Multi-cell models accurately predicted chemokine and cytokine profiles in simulated inflamed and cancerous tissues.
• Model predictions for gingival epithelial keratinocytes, dendritic cells, and T lymphocytes showed high concordance with experimental results (75-80% accuracy).
• Personalized multi-cell models incorporating genomic data accurately predicted multiple myeloma cell line responses (75% for cytokine profiles, 100% for marker expression).

Conclusions:

• Multi-cell computational models offer a validated approach to predict complex cellular interactions and biomarker profiles in disease.
• These models can be personalized using genomic data for more accurate disease simulation.
• The models show potential as high-throughput screening tools for anti-inflammatory and immuno-oncology drug discovery.