Characterizing stutter variants in forensic STRs with massively parallel sequencing. (March 2020)

• Record Type: Journal Article
• Authors: Li, Ran; Wu, Riga; Li, Haixia; Zhang, Yinming; Peng, Dan; Wang, Nana; Shen, Xuefeng; Wang, Zhiyuan; Sun, Hongyu
• Is Part Of: Forensic science international. Volume 45(2020)
• Publication Date: 2020-03
• Subjects: Massively parallel sequencing (MPS) -- Short tandem repeat (STR) -- Stutter; Forensic genetics -- Periodicals; Génétique légale -- Périodiques; Forensic genetics; Electronic journals; Periodicals; 614.1
• DOI: 10.1016/j.fsigen.2019.102225
• Languages: English
• ISSNs: 1872-4973

Abstract

Highlights: A sequence simplification strategy was introduced. More stutter variants—including N-4, N-3, N-2, N0, N + 1, and N + 2—were studied. One-repeat-unit-longer stutters (or PAs) were better predictors for backward stutters. Co-stuttering patterns were observed between two adjacent continuous motifs. Abstract: Despite improvements in characterizing stutters of short tandem repeats (STRs), the relationships among the amounts of stutter variants and the relationships among motifs are not well understood yet. In the present study, 750 peripheral blood samples from human subjects were included to characterize the stutters of 58 STRs via the ForenSeq DNA Signature Prep Kit on a MiSeq FGx instrument. Alleles and corresponding stutter products were identified with a sequence simplification procedure. After screening, 26, 921 alleles were included, that resulted in over 50 million reads, among which 8.69% were stutter products. Among these stutter products, 83.44% were N-1 stutters. Additionally, N-4, N-3, N-2, N0, N + 1, and N + 2 variants accounted for 0.11%, 0.77%, 6.45%, 3.01%, 5.95%, and 0.25% of the stutter products, respectively. For backward stutters, stutter products correlated best with the corresponding one-unit-longer stutter (or parental allele), which may represent a good predictor for backward stutters. For forward stutters, the N + 2 stutter correlated best with the N + 1 stutter, whereas the N + 1 stutter correlated best with the N-1 stutter rather than the expected parental allele, which indicated that the patterns were more complex for forward stutters. Additionally, some interesting findings were obtained for D21S11. For two adjacent contiguous motifs, co-stuttering patterns were observed where one motif tended to increase one repeat unit while the other motif decreased one repeat unit, whereas the inter-motif dependency was not significant for interrupted motifs. In conclusion, with massively parallel sequencing technology and our sequence simplification strategy, sequence variations within alleles and stutter products were identified, which was useful to determine the origin of stutters, identify more stutter variants, and explore the relationships among motifs. These findings may be helpful for allele designation, a deeper understanding of the mechanism of stutter, and improving resolution in forensic mixture analyses.