Date of this Version
The polymerase chain reaction (PCR) is continually growing in its application, particularly in the field of molecular diagnosis of disease from clinical specimens. The main focus has been in the detection and identification of pathogens. However, quantitative PCR is increasingly utilized to determine initial pathogen load. A well-designed PCR protocol is required in all of these instances. Just as importantly, in the context of disease diagnosis; is the design of the sample processing methodology. The ideal method should concentrate the DNA and effectively isolate a high-quality DNA product, free of PCR inhibitors, while also being simple, reproducible and safe.
The aim of this work is to address the research challenges posed in the preceding paragraph. A previously developed prototype diagnostic system is used to analyze and suggest improvements and an application of the technology is also described. Briefly, the system includes a polystyrene strip that is inserted into a lysis microreactor (LMR) that is fitted with an impeller and temperature control to lyse DNA. The DNA binds noncovalently to the strip and is transferred through a wash step to the thermocycler cuvette for amplification.
The research challenges were addressed by the following:
- An analytical model was developed to determine the efficiency of each process comprising a PCR cycle. Using this model, reaction conditions can be directly linked to the overall yield and initial template concentration can be determined from real-time PCR data.
- The flow characteristic of the LMR was solved by computational fluid dynamics to determine the DNA capture efficiency as a function of initial position.
- Improvements to the use of a non-specific strip for DNA binding were explored by attaching target-complimentary oligonucleotides to a surface.
- The prototype system was evaluated on a bank of frozen clinical stool samples. Samples were tested for Clostridium difficile genomic DNA and the results compared with standard C. difficile testing methods used routinely by a hospital clinical laboratory. The prototype system showed 97.5% concordance with standard testing methods.
Adviser: Hendrik J. Viljoen