Using mechanistic models and machine learning to design single-color multiplexed nascent chain tracking experiments.

Area of Science:

• Molecular Biology
• Cell Biology
• Biophysics

Background:

• mRNA translation is a fundamental cellular process, crucial for protein synthesis.
• Nascent chain tracking (NCT) enables single-molecule resolution of mRNA translation dynamics in live cells.
• Current NCT methods are limited to observing only one or two mRNA species simultaneously due to fluorescent tag limitations.

Purpose of the Study:

• To develop a computational strategy for enhancing the multiplexing capability of Nascent chain tracking (NCT).
• To enable simultaneous observation of multiple mRNA species within a single cell using advanced imaging techniques.
• To explore new experimental designs for studying complex cellular processes like cell signaling.

Main Methods:

• A hybrid computational pipeline integrating detailed mechanistic simulations with machine learning (ML) was developed.
• Simulations generated realistic NCT videos to assess experimental designs.
• ML algorithms were employed to evaluate the potential for resolving multiple mRNA species using limited fluorescent tags.

Main Results:

• The proposed hybrid strategy demonstrated the potential to significantly increase the number of simultaneously trackable mRNA species.
• A simulated NCT experiment successfully identified seven different mRNA species within a single cell using only two fluorescent tags.
• The ML-based labeling achieved 90% accuracy in identifying individual mRNA translation spots.

Conclusions:

• The developed computational approach offers a viable method to extend the multiplexing capacity of NCT.
• This advancement allows for simultaneous monitoring of multiple mRNAs, opening new avenues in live-cell imaging.
• The findings are particularly relevant for cell signaling studies requiring the concurrent analysis of diverse mRNA dynamics.