Library construction optimization and analysis of Short Tandem Repeats by a simple, PCR-based DNA barcoding method

Sammanfattning: DNA profiling is evolving in the forensics community towards introducing massively parallel sequencing (MPS) as a complement to capillary electrophoresis (CE). Obstacles remain however before this technology can become routine in forensic casework, such as the development of more efficient bioinformatics solutions and recommendations concerning data analyses. Additionally, there is a need to develop MPS methods that are even more sensitive than current commercialized systems. One promising candidate is SiMSen-Seq, a method that incorporates unique molecular identifiers (UMI’s), also known as barcodes, into PCR library preparation, allowing for the reduction of background noise (artefacts) in data analyses. In the development of SiMSen-Seq towards its use in forensics, the library preparation protocol, consisting of two distinct PCRs (PCR1 and PCR2), were in this project further optimized for the efficient amplification of short tandem repeats (STRs). By applying Bioanalyzer 2100, the results showed that the type of DNA polymerase and the barcode primer concentration had the greatest effect in maximizing specific products and minimizing nonspecific products. SuperFi and Immolase, two promising DNA polymerases that resulted in efficient STR amplification, were further evaluated for use in library preparation by MPS using MiSeq, and the results showed that although the use of SuperFi in PCR1 and Immolase in PCR2 resulted in the most STR products, it generated the highest amount of artefacts, complicating data interpretation. Instead, utilizing SuperFi, a proofreading enzyme, in PCR2 of library preparation, decreased the amount of generated artefacts. Based on these results, it is therefore recommended that further tests are performed with SuperFi in both PCR1&2 of library preparation. Testing other DNA polymerase combinations featuring proofreading abilities may also provide valuable data. However, before SiMSen-Seq can be implemented in the analyses of real crime scene samples, additional evaluation using low DNA concentrations and more complex DNA samples, including inhibitors, is required. Further customization of the bioinformatic data workflow is also necessary to streamline the work process.