Developing a Machine Learning 'Smart' Polymerase Chain Reaction Thermocycler Part 2: Putting the Theoretical...

Area of Science:

• Forensic Biology
• Molecular Biology
• Biotechnology

Background:

• Polymerase Chain Reaction (PCR) has revolutionized forensic science, enhancing DNA profiling sensitivity and discrimination.
• Challenges persist in DNA profiling of trace, inhibited, and degraded samples.
• Optimizing PCR performance for sub-optimal samples is crucial for forensic applications.

Purpose of the Study:

• To develop a proof-of-concept for a smart PCR system capable of optimizing DNA profiling.
• To investigate the effects of altering PCR cycling conditions on DNA profile quality.
• To reduce PCR runtime while maintaining acceptable DNA product quality and quantity.

Main Methods:

• Manual alteration of denaturation and annealing temperatures and timings in PCR.
• Trial of a real-time feedback system using STR PCR and qPCR.
• Comparison of DNA profiles generated by the modified PCR with standard STR PCR kits.
• Exploration of machine learning for real-time PCR adjustments.

Main Results:

• Identified PCR cycling parameters that produced results comparable to standard endpoint PCR.
• Achieved a reduction in PCR runtime by 30 minutes.
• Demonstrated the feasibility of a smart PCR system for optimizing DNA profiling.

Conclusions:

• Altering PCR cycling conditions can optimize performance for challenging DNA samples.
• A smart PCR system, potentially leveraging machine learning, can enable real-time adjustments for predefined goals.
• This technology has significant implications for various biological disciplines reliant on PCR.