Experts at Flinders University predict exciting developments in essential DNA testing through the integration of machine learning into DNA profiling. PCR (polymerase chain reaction) DNA profiling has transformed high-throughput sampling in various fields including medical diagnostics, forensic assessments, and national security this century. However, the method itself has seen little innovation since it was created in the 1980s.
Experts from Flinders University foresee significant advancements in vital DNA testing thanks to the application of machine learning in DNA profiling.
Over the last century, PCR (polymerase chain reaction) DNA profiling has dramatically changed high-throughput sampling for medical diagnostics, forensic investigations, and national security. Despite this shift, the core techniques have remained largely unchanged since the 1980s.
“Even a minor enhancement in PCR efficiency could greatly influence the hundreds of thousands of forensic and intelligence DNA samples processed annually, especially when dealing with degraded samples,” note experts, including Dr. Duncan Taylor from Forensic Science SA at Flinders University.
Recent research published in Genes highlights substantial improvements in DNA profiling quality and more efficient PCR cycling methods through the use of artificial intelligence techniques. The study was led by College of Science and Engineering PhD student Ms. Caitlin McDonald.
“Our approach can tackle issues that have plagued forensic scientists for many years, particularly regarding trace, inhibited, or degraded samples,” McDonald remarked after presenting the findings at the International Society of Forensic Genetics conference.
“By smartly optimizing PCR for a diverse range of sample types, we can significantly improve amplification success rates, leading to more dependable results even in challenging scenarios.”
“In addition to forensic applications, this technology could transform other domains reliant on PCR, such as clinical diagnostics and environmental monitoring, by enhancing efficiency, minimizing errors, and allowing for large-scale analyses across various uses.”
PCR is a commonly used laboratory procedure that amplifies small strands of genetic material; it has applications in DNA fingerprinting, diagnosing genetic disorders, and identifying pathogens like COVID-19.
Supported by fellow experts from Flinders University’s College of Science and Engineering, including Professor Adrian Linacre and AI computer scientist Associate Professor Russell Brinkworth, the study employed machine learning to develop innovative ‘smart PCR’ systems. These systems are designed to identify significant adjustments and achieve faster cycling times for quicker and more precise results.
The article published in Genes elaborates on the development and extensive testing of the newly conceived ‘smart’ PCR system.
Professor Linacre, who specializes in DNA forensic developments, states that PCR is utilized across multiple sectors, including forensic science, veterinary research, medical practices, and national safety.
“Artificial intelligence and machine learning are still in their early stages but have the potential, when leveraged appropriately, to greatly enhance the sensitivity of PCR testing,” explains Professor Linacre.
He notes that investigations of non-coding DNA sections for forensic purposes have been underway since 1994.
“With continued research, these AI-driven methods can potentially elevate the standard of DNA evidence used in criminal investigations and improve the quality of trace DNA samples, thereby advancing the criminal justice system.”
Associate Professor of Autonomous Systems Russell Brinkworth emphasizes that enhancements to existing protocols will further shape the future of AI implementations.
“Traditionally, DNA amplification required all variables to be set before initiating the process, which didn’t account for the various sample and condition differences,” states Associate Professor Brinkworth.
“By leveraging advancements in machine learning and sensor technology, we have transitioned the PCR process from a generic method to one that is tailored and optimized for individual cases, enabling faster production of higher quality and quantity DNA than ever before.”
