Comparative performance of the BGISEQ-500 vs Illumina HiSeq2500 sequencing platforms for palaeogenomic sequencing

Mak, Sarah Siu Tze; Gopalakrishnan, Shyam; Carøe, Christian; Geng, Chunyu; Liu, Shanlin; Sinding, Mikkel Holger S.; Kuderna, Lukas, 1989-; Zhang, Wenwei; Fu, Shujin; Vieira, Filipe Garrett; Germonpré, Mietje; Bocherens, Hervé; Fedorov, Sergey E.; Petersen, Bent; Sicheritz-Ponten, Thomas; Marquès i Bonet, Tomàs, 1975-; Zhang, Guojie; Jiang, Hui; Gilbert, M Thomas

doi:http://dx.doi.org/10.1093/gigascience/gix049

Comparative performance of the BGISEQ-500 vs Illumina HiSeq2500 sequencing platforms for palaeogenomic sequencing

Citation

Mak SST, Gopalakrishnan S, Carøe C, Geng C, Liu S, Sinding MS et al. Comparative performance of the BGISEQ-500 vs Illumina HiSeq2500 sequencing platforms for palaeogenomic sequencing. Gigascience. 2017 Aug 1;6(8):1-13. DOI: 10.1093/gigascience/gix049

Abstract

Ancient DNA research has been revolutionized following development of next-generation sequencing platforms. Although a number of such platforms have been applied to ancient DNA samples, the Illumina series are the dominant choice today, mainly because of high production capacities and short read production. Recently a potentially attractive alternative platform for palaeogenomic data generation has been developed, the BGISEQ-500, whose sequence output are comparable with the Illumina series. In this study, we modified the standard BGISEQ-500 library preparation specifically for use on degraded DNA, then directly compared the sequencing performance and data quality of the BGISEQ-500 to the Illumina HiSeq2500 platform on DNA extracted from 8 historic and ancient dog and wolf samples. The data generated were largely comparable between sequencing platforms, with no statistically significant difference observed for parameters including level (P = 0.371) and average sequence length (P = 0718) of endogenous nuclear DNA, sequence GC content (P = 0.311), double-stranded DNA damage rate (v. 0.309), and sequence clonality (P = 0.093). Small significant differences were found in single-strand DNA damage rate (δS; slightly lower for the BGISEQ-500, P = 0.011) and the background rate of difference from the reference genome (θ; slightly higher for BGISEQ-500, P = 0.012). This may result from the differences in amplification cycles used to polymerase chain reaction-amplify the libraries. A significant difference was also observed in the mitochondrial DNA percentages recovered (P = 0.018), although we believe this is likely a stochastic effect relating to the extremely low levels of mitochondria that were sequenced from 3 of the samples with overall very low levels of endogenous DNA. Although we acknowledge that our analyses were limited to animal material, our observations suggest that the BGISEQ-500 holds the potential to represent a valid and potentially valuable alternative platform for palaeogenomic data generation that is worthy of future exploration by those interested in the sequencing and analysis of degraded DNA.