VIP-OT: Dissecting Single-Cell Biochemical State Dynamics under Perturbation via Vibrational Painting and Optimal...

Area of Science:

• Single-cell analysis
• Biophysics
• Computational biology

Background:

• Understanding cellular heterogeneity in response to perturbations is crucial for cell biology.
• Current methods struggle to model dynamic cell state transitions.
• Vibrational imaging offers label-free molecular fingerprints but lacks dynamic analysis capabilities.

Purpose of the Study:

• To introduce Vibrational Painting-Optimal Transport (VIP-OT), an integrated framework for analyzing single-cell perturbation responses.
• To computationally reconstruct cell state transition trajectories from static imaging data.
• To enable predictive modeling of cellular responses to chemical treatments.

Main Methods:

• Coupling multiplexed infrared (IR) and Raman imaging with optimal transport algorithms.
• Analyzing over 22,000 single-cell spectra from human breast adenocarcinoma cells under 16 drug treatments.
• Developing a machine learning model for predicting post-treatment metabolic states from pre-treatment spectra.

Main Results:

• VIP-OT successfully traced drug response heterogeneity back to baseline metabolic states.
• The machine learning model accurately predicted individual cell metabolic states post-treatment.
• The concept of Spectral Velocity was introduced to map dynamic response trajectories and resolve drug combination effects.

Conclusions:

• VIP-OT provides a novel framework for dissecting heterogeneous cellular responses at single-cell resolution.
• This approach enables the development of virtual cell simulators for perturbation studies.
• VIP-OT leverages high-content vibrational imaging and interpretable computational modeling for dynamic cell analysis.