Andreas Halman1,2,3,4, Alicia Oshlack1,2,5. 1. Murdoch Children's Research Institute, Royal Children's Hospital, Parkville, VIC, 3052, Australia. 2. Peter MacCallum Cancer Centre, 305 Grattan St, Melbourne, VIC, 3000, Australia. 3. Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC, 3052, Australia. 4. School of Natural Sciences and Health, Tallinn University, Tallinn, 10120, Estonia. 5. School of BioSciences, University of Melbourne, Parkville, VIC, 3052, Australia.
Abstract
Background: Short tandem repeats are an important source of genetic variation. They are highly mutable and repeat expansions are associated dozens of human disorders, such as Huntington's disease and spinocerebellar ataxias. Technical advantages in sequencing technology have made it possible to analyse these repeats at large scale; however, accurate genotyping is still a challenging task. We compared four different short tandem repeats genotyping tools on whole exome sequencing data to determine their genotyping performance and limits, which will aid other researchers in choosing a suitable tool and parameters for analysis. Methods: The analysis was performed on the Simons Simplex Collection dataset, where we used a novel method of evaluation with accuracy determined by the rate of homozygous calls on the X chromosome of male samples. In total we analysed 433 samples and around a million genotypes for evaluating tools on whole exome sequencing data. Results: We determined a relatively good performance of all tools when genotyping repeats of 3-6 bp in length, which could be improved with coverage and quality score filtering. However, genotyping homopolymers was challenging for all tools and a high error rate was present across different thresholds of coverage and quality scores. Interestingly, dinucleotide repeats displayed a high error rate as well, which was found to be mainly caused by the AC/TG repeats. Overall, LobSTR was able to make the most calls and was also the fastest tool, while RepeatSeq and HipSTR exhibited the lowest heterozygous error rate at low coverage. Conclusions: All tools have different strengths and weaknesses and the choice may depend on the application. In this analysis we demonstrated the effect of using different filtering parameters and offered recommendations based on the trade-off between the best accuracy of genotyping and the highest number of calls. Copyright:
Background: Short tandem repeats are an important source of genetic variation. They are highly mutable and repeat expansions are associated dozens of human disorders, such as Huntington's disease and spinocerebellar ataxias. Technical advantages in sequencing technology have made it possible to analyse these repeats at large scale; however, accurate genotyping is still a challenging task. We compared four different short tandem repeats genotyping tools on whole exome sequencing data to determine their genotyping performance and limits, which will aid other researchers in choosing a suitable tool and parameters for analysis. Methods: The analysis was performed on the Simons Simplex Collection dataset, where we used a novel method of evaluation with accuracy determined by the rate of homozygous calls on the X chromosome of male samples. In total we analysed 433 samples and around a million genotypes for evaluating tools on whole exome sequencing data. Results: We determined a relatively good performance of all tools when genotyping repeats of 3-6 bp in length, which could be improved with coverage and quality score filtering. However, genotyping homopolymers was challenging for all tools and a high error rate was present across different thresholds of coverage and quality scores. Interestingly, dinucleotide repeats displayed a high error rate as well, which was found to be mainly caused by the AC/TG repeats. Overall, LobSTR was able to make the most calls and was also the fastest tool, while RepeatSeq and HipSTR exhibited the lowest heterozygous error rate at low coverage. Conclusions: All tools have different strengths and weaknesses and the choice may depend on the application. In this analysis we demonstrated the effect of using different filtering parameters and offered recommendations based on the trade-off between the best accuracy of genotyping and the highest number of calls. Copyright:
Authors: Rick M Tankard; Mark F Bennett; Peter Degorski; Martin B Delatycki; Paul J Lockhart; Melanie Bahlo Journal: Am J Hum Genet Date: 2018-11-29 Impact factor: 11.025
Authors: Mark A DePristo; Eric Banks; Ryan Poplin; Kiran V Garimella; Jared R Maguire; Christopher Hartl; Anthony A Philippakis; Guillermo del Angel; Manuel A Rivas; Matt Hanna; Aaron McKenna; Tim J Fennell; Andrew M Kernytsky; Andrey Y Sivachenko; Kristian Cibulskis; Stacey B Gabriel; David Altshuler; Mark J Daly Journal: Nat Genet Date: 2011-04-10 Impact factor: 38.330
Authors: Egor Dolzhenko; Joke J F A van Vugt; Richard J Shaw; Mitchell A Bekritsky; Marka van Blitterswijk; Giuseppe Narzisi; Subramanian S Ajay; Vani Rajan; Bryan R Lajoie; Nathan H Johnson; Zoya Kingsbury; Sean J Humphray; Raymond D Schellevis; William J Brands; Matt Baker; Rosa Rademakers; Maarten Kooyman; Gijs H P Tazelaar; Michael A van Es; Russell McLaughlin; William Sproviero; Aleksey Shatunov; Ashley Jones; Ahmad Al Khleifat; Alan Pittman; Sarah Morgan; Orla Hardiman; Ammar Al-Chalabi; Chris Shaw; Bradley Smith; Edmund J Neo; Karen Morrison; Pamela J Shaw; Catherine Reeves; Lara Winterkorn; Nancy S Wexler; David E Housman; Christopher W Ng; Alina L Li; Ryan J Taft; Leonard H van den Berg; David R Bentley; Jan H Veldink; Michael A Eberle Journal: Genome Res Date: 2017-09-08 Impact factor: 9.438