Poor-quality samples are common in forensic casework. Analysis of such samples can be challenging as DNA is often degraded into small fragments which can hinder successful amplification of target regions. Single nucleotide polymorphisms (SNPs) have greater potential to generate useful data from degraded samples, in comparison to traditional short tandem repeat (STR) analysis, due to their smaller amplicon size. Analysis of these markers can be particularly beneficial for missing person identification when using familial references, where targeting a large number of SNP loci can allow for the identification of a range of genetic relationships with high levels of confidence. However, reliable kinship likelihood ratios for such an application are greatly dependent on accurate SNP profiles.
ForenSeq® Kintelligence is a large-scale SNP panel that includes 10,230 markers in a single assay. Published work has shown that the original configuration of the Kintelligence kit has been used to analyse poor-quality samples with success. The newer high throughput (HT) configuration of this assay is designed to allow samples to be sequenced at a higher plexity, meaning that more libraries can be analysed on one run.
One of the most important factors influencing how many sample libraries can be sequenced simultaneously is the proportion of reads that are lost to off-target DNA sequences. These are most often adaptor-adaptor or primer-adaptor dimers which will also bind to the sequencing flow cell. Both configurations of the Kintelligence kit use a simple library preparation workflow, in which bead-based cleanups that are performed after target amplification and index adaptor ligation in order to remove any unused primers, index adaptors, and other artefacts from the libraries.
Initial work with Kintelligence HT has highlighted that when libraries are prepared from poor-quality samples, a higher proportion of off-target reads can be observed in comparison to libraries prepared from good quality samples. Having such a high proportion of off-target reads will ultimately reduce sequencing capacity for target libraries and limit sample plexity.
This study aimed to increase the number of on-target reads when analysing poor quality samples by modifying the Kintelligence workflow. Modification included testing different cleanup methods at various stages of library preparation, and sequencing libraries at different levels of plexity. The results from this work showed that the modified protocol substantially improved sequencing success from poor quality samples.