Enter Note Done

Session


Keywords: Exome sequencing; Genetic testing; Genome sequencing; Mutation detection; Targeted sequencing

Authors:
F. Mafra 1; J. Garifallou 1; L. Kulikoski 2; A. Dias 2; D. Li 1; J. Smiler 1; N. Williams 1; A. Caro 1; C. Vaccaro 1; H. Hakonarson 1,3; R. Pellegrino 1

Affiliations:
1) Center for Applied Genomics, The Childrens Hospital of Philadelphia, Philadelphia, Pennsylvania.; 2) USP, Universidade de Sao Paulo, Pathology Division.; 3) Department of Pediatrics, The Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA


Introduction: Sequencing DNA extracted from dry blood spots and formalin-fixed and paraffin-embedded specimens (FFPE) can frequently provide limited amount of degraded, poor quality DNA. The inconsistent input volume and DNA damage, such as inter and intra strand crosslinks, accumulation of single-stranded DNA, and single-strand DNA breaks, prevalent in these types of samples can both negatively affect sequencing results. The effects can range from simple library failures to generating libraries that produce unauthentic data, leading to misinterpretation of the results. To overcome these issues, we evaluated the performance of the Twist Library Preparation Kit in poor quality samples.

Methods: We tested a total of fifty-two challenging samples. Amongst the cohort, there were two poor quality gDNA (saliva), four FFPE, and forty-six dry blood spots. Samples were manually processed to test the Twist Library Preparation Kit that used enzymatic fragmentation and Y-shaped adapters to generate dual indexed libraries, followed by target enrichment using Twist capture probes and hybridization reagents. Sequencing was performed on an Illumina HiSeq 2500 sequencer in Rapid Run mode and the bioinformatics analysis was completed using Illumina DRAGEN Bio-IT Software. We targeted 60X mean coverage for the two poor quality gDNA and four FFPE samples using the Twist Human Core Exome kit. In addition, we targeted 200X mean coverage for the forty-six dry blood spots using an in-house designed Twist custom epilepsy panel.

Results: Captured libraries were successfully generated according to the Center for Applied Genomics (CAG) quality standards/metrics for all the samples in question. The assay performance was evaluated by checking the quality of sequencing (passing filters), duplicate rates, on target, and coverage at bases on 20x; particularly by looking at the number of reads and the amount of data generated and by determining the presence of good coverage and high uniformity across the curated genes of interest.

Conclusions: The kit tested was able to overcome challenges in DNA purification, including low yields, poor quality, and difficult sample types. Following the initial tests, the Center for Applied Genomics NGS unit established the Twist Core Exome pipeline, and we are now optimizing the procedures on PerkinElmer Sciclone instruments for high-throughput use of the kit. Further improvements are still in development.